{ "complete": true, "report": { "bucket": "emo_Confusion", "label": "Intense Confusion", "definition": { "id": "emo_Confusion", "category": "emotion", "head": "emo_Confusion", "pole": "high", "label": "Intense Confusion", "floor": 3.0 }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/emotion_confusion/full-v2-sealed/sealed-selected", "mission": "emotion_confusion", "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": 6.226780033111572, "genuineness": 4.508351287841797, "wer": 0.013974321411612846, "identity": 0.6184235051274299, "voiceclap": -3.475375473499298e-05, "fitness": 0.7470230198557278, "steering_v2": 0.8285993467916132, "raw_target": 0.9878124940395355, "target_cdf": 0.8285993467916132, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.223166103363037, "genuineness": 4.4917825412750245, "wer": 0.015307444440091826, "identity": 0.6184235051274299, "fitness": 0.753781327293985, "steering_v2": 0.8459182500186971 }, "none": { "n": 50, "ce": 6.2303939628601075, "genuineness": 4.524920034408569, "wer": 0.012641198383133868, "identity": null, "fitness": 0.7402647124174707, "steering_v2": 0.8112804435645292 } } } } ], "selected_wording": { "id": "manual-de7bf3e68eea", "general": "Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.", "inline": [ "Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.", "Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.", "The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization." ], "cfg": 3, "recipe": "P05", "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": 5.0 }, "n": 20, "target_z": 2.0269531195437445, "target_raw": 1.3989364758133889, "ce": 5.623494362831115, "genuineness": 4.388102340698242, "identity": 0.5217329245060682, "wer": 0.049743077055383556, "valid_fraction": 1, "identity_delta": -0.04934858791530132, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 2.011962900506279, "target_raw": 1.390940648317337, "ce": 5.563154935836792, "genuineness": 4.357977390289307, "identity": 0.5680798128247261, "wer": 0.02402046783625731, "valid_fraction": 1, "identity_delta": -0.003001699596643448, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.0269531195437445, "target_raw": 1.3989364758133889, "ce": 5.623494362831115, "genuineness": 4.388102340698242, "identity": 0.5217329245060682, "wer": 0.049743077055383556, "valid_fraction": 1, "identity_delta": -0.04934858791530132, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.2084716287833626, "target_raw": 0.9623560100793839, "ce": 5.8585974216461185, "genuineness": 4.478477334976196, "identity": 0.5537032060325145, "wer": 0.01112831097351221, "valid_fraction": 1, "identity_delta": -0.01737830638885498, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.9050537394218828, "target_raw": 0.8005120038986206, "ce": 5.978531789779663, "genuineness": 4.473770236968994, "identity": 0.5983803480863571, "wer": 0.013797514619883041, "valid_fraction": 1, "identity_delta": 0.027298835664987565, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.4292724706523633, "target_raw": 0.5467288576066494, "ce": 5.878652715682984, "genuineness": 4.449607563018799, "identity": 0.562470567971468, "wer": 0.007894736842105263, "valid_fraction": 1, "identity_delta": -0.00861094444990158, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 0.27480470201364016, "target_raw": 0.4643352895975113, "ce": 6.085956382751465, "genuineness": 4.513936924934387, "identity": 0.6161166042089462, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04503509178757668, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 0.05415040137927998, "target_raw": 0.3466376282274723, "ce": 6.101841783523559, "genuineness": 4.496677803993225, "identity": 0.5980301514267922, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": 0.026948639005422593, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.06008299989820111, "target_raw": 0.285705191642046, "ce": 6.081190800666809, "genuineness": 4.441762518882752, "identity": 0.5847143813967705, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.013632868975400924, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.7126220874315021, "target_raw": 1.231271366775036, "ce": 5.698949766159058, "genuineness": 4.384650540351868, "identity": 0.5307223200798035, "wer": 0.037884201926384586, "valid_fraction": 0.85, "identity_delta": -0.04035919234156608, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 2.011962900506279, "target_raw": 1.390940648317337, "ce": 5.563154935836792, "genuineness": 4.357977390289307, "identity": 0.5680798128247261, "wer": 0.02402046783625731, "valid_fraction": 1, "identity_delta": -0.003001699596643448, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.9847656494003754, "target_raw": 1.3764335535466672, "ce": 5.532054424285889, "genuineness": 4.332873153686523, "identity": 0.5646109640598297, "wer": 0.05008040935672514, "valid_fraction": 0.9, "identity_delta": -0.006470548361539841, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.7548828198113826, "target_raw": 1.2538133673369884, "ce": 5.61297037601471, "genuineness": 4.3906127452850345, "identity": 0.5563407938927412, "wer": 0.02496345029239766, "valid_fraction": 1, "identity_delta": -0.01474071852862835, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.553759749534502, "target_raw": 1.1465337224304677, "ce": 5.68922004699707, "genuineness": 4.443645286560058, "identity": 0.5409371323883534, "wer": 0.019353070175438594, "valid_fraction": 1, "identity_delta": -0.030144380033016206, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.2136719024558786, "target_raw": 0.9651298515498639, "ce": 5.854626083374024, "genuineness": 4.4907156229019165, "identity": 0.5858170714229345, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.01473555900156498, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.170166044494049, "target_raw": 0.9419236972928047, "ce": 5.858597445487976, "genuineness": 4.481615304946899, "identity": 0.5798789292573929, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.008797416836023331, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.0561401087404347, "target_raw": 0.8811019234359264, "ce": 5.7624911069869995, "genuineness": 4.42387580871582, "identity": 0.5812010653316975, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.01011955291032791, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 0.7208251839280225, "target_raw": 0.7022439435124397, "ce": 5.9249187231063845, "genuineness": 4.526488995552063, "identity": 0.5455016575753688, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.02557985484600067, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.634179700966039, "target_raw": 0.6560269847512246, "ce": 5.948548221588135, "genuineness": 4.48757758140564, "identity": 0.594646880030632, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.023565367609262467, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.608618172811606, "target_raw": 0.6423923894762993, "ce": 5.908239150047303, "genuineness": 4.444272971153259, "identity": 0.5835311368107796, "wer": 0.008742690058479531, "valid_fraction": 1, "identity_delta": 0.01244962438941002, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.5704101718658038, "target_raw": 0.6220121279358863, "ce": 6.011890983581543, "genuineness": 4.473142695426941, "identity": 0.5959809720516205, "wer": 0.008596491228070175, "valid_fraction": 1, "identity_delta": 0.02489945963025093, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.34897462481074737, "target_raw": 0.5038977473974228, "ce": 5.984091663360596, "genuineness": 4.513623142242432, "identity": 0.56108987480402, "wer": 0.01111111111111111, "valid_fraction": 1, "identity_delta": -0.009991637617349624, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.28803710927133114, "target_raw": 0.4713934950530529, "ce": 5.9459668636322025, "genuineness": 4.463728666305542, "identity": 0.5772552281618119, "wer": 0.014215686274509804, "valid_fraction": 1, "identity_delta": 0.006173715740442276, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 0.22436524173548894, "target_raw": 0.43743073120713233, "ce": 6.035321831703186, "genuineness": 4.496050214767456, "identity": 0.565158686041832, "wer": 0.008660130718954249, "valid_fraction": 1, "identity_delta": -0.005922826379537583, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 0.1329101551041863, "target_raw": 0.38864831551909446, "ce": 6.104820203781128, "genuineness": 4.552220749855041, "identity": 0.5327463716268539, "wer": 0.008680555555555556, "valid_fraction": 1, "identity_delta": -0.03833514079451561, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 0.03811036871423978, "target_raw": 0.3380818270146847, "ce": 6.1072030544281, "genuineness": 4.472201287746429, "identity": 0.5854119896888733, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.014330477267503739, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 0.022875955258549626, "target_raw": 0.3299557454884052, "ce": 6.093303322792053, "genuineness": 4.386847138404846, "identity": 0.5212876766920089, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.04979383572936058, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 0.002416980546578467, "target_raw": 0.31904286742210386, "ce": 6.080396509170532, "genuineness": 4.513623118400574, "identity": 0.5538860842585563, "wer": 0.008742690058479531, "valid_fraction": 1, "identity_delta": -0.017195428162813185, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.0016601446281753675, "target_raw": 0.31686811670660975, "ce": 6.052994298934936, "genuineness": 4.509229946136474, "identity": 0.6091802835464477, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": 0.0380987711250782, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.06118163872076675, "target_raw": 0.28511917442083357, "ce": 6.050412964820862, "genuineness": 4.508288478851318, "identity": 0.5785847947001457, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.007503282278776169, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.12138671627285992, "target_raw": 0.2530056066811085, "ce": 6.114748597145081, "genuineness": 4.429524207115174, "identity": 0.5790344417095185, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.00795292928814888, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.12277832163018576, "target_raw": 0.25226332023739817, "ce": 6.1095858335495, "genuineness": 4.4301518201828, "identity": 0.560432356595993, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.01064915582537651, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.1248046740776108, "target_raw": 1.451130786538124, "ce": 5.493929553031921, "genuineness": 4.310593247413635, "identity": 0.5092191495001316, "wer": 0.055051169590643274, "valid_fraction": 0.9, "identity_delta": -0.061862362921237944, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.395458980299177, "target_raw": 1.06209562048316, "ce": 5.734393954277039, "genuineness": 4.4587077617645265, "identity": 0.5306991949677468, "wer": 0.027777777777777776, "valid_fraction": 0.85, "identity_delta": -0.04038231745362282, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.2969238241473896, "target_raw": 0.15937359035015106, "ce": 6.169155907630921, "genuineness": 4.363939559459686, "identity": 0.5710815124213695, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": -0.41215820407984477, "target_raw": 0.09790722876787186, "ce": 6.191395378112793, "genuineness": 4.350759887695313, "identity": 0.5994429692626, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": 0.02836145684123039, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -0.40419922098706484, "target_raw": 0.10215257406234741, "ce": 6.235080122947693, "genuineness": 4.363939547538758, "identity": 0.5684736454859376, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.002607866935431957, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.45410156500987764, "target_raw": 0.07553451508283615, "ce": 6.203309392929077, "genuineness": 4.276388871669769, "identity": 0.5819923400878906, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.010910827666521072, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.382031247119097, "target_raw": 0.11397703737020493, "ce": 6.265818223953247, "genuineness": 4.348563270568848, "identity": 0.591278854906559, "wer": 0.010263504611330698, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.9349218852752208, "target_raw": 1.349846741259098, "ce": 5.695177001953125, "genuineness": 4.35534143447876, "identity": 0.547153008878231, "wer": 0.05000473515805323, "valid_fraction": 0.88, "identity_delta": -0.04412584602832794, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 2.057070302246604, "target_raw": 1.4150010707974434, "ce": 5.722380638122559, "genuineness": 4.359609141349792, "identity": 0.5283753668516874, "wer": 0.08755632658721904, "valid_fraction": 0.92, "identity_delta": -0.06290348805487156, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.9349218852752208, "target_raw": 1.349846741259098, "ce": 5.695177001953125, "genuineness": 4.35534143447876, "identity": 0.547153008878231, "wer": 0.05000473515805323, "valid_fraction": 0.88, "identity_delta": -0.04412584602832794, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.12214841697138358, "ci95_z": [ -0.08992190036067646, 0.3342187343034436 ], "language_deltas": { "en": 0.4068359100222963, "de": 0.00429686578984041, "fr": -0.057812528459829704, "nl": 0.4265624928413925, "es": -0.16914065533678163 }, "languages_improved": 3, "clear_gain": false, "delta_ce": 0.027203636169433595, "delta_genuineness": 0.004267706871032715, "delta_wer": 0.0375515914291658, "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": 0.12214841697138358, "ci95_z": [ -0.08992190036067646, 0.3342187343034436 ], "language_deltas": { "en": 0.4068359100222963, "de": 0.00429686578984041, "fr": -0.057812528459829704, "nl": 0.4265624928413925, "es": -0.16914065533678163 }, "languages_improved": 3, "clear_gain": false, "delta_ce": 0.027203636169433595, "delta_genuineness": 0.004267706871032715, "delta_wer": 0.0375515914291658, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 2.3169531323943175, "ci95_z": [ 1.5715625129553334, 2.8352343821673376 ], "language_deltas": { "en": 0.9083984447677326, "de": 2.77460938447206, "fr": 2.203710960656349, "nl": 2.7691406022583265, "es": 2.9289062698171207 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.570641222000122, "delta_genuineness": 0.0067781639099121094, "delta_wer": 0.03974123054672254, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 1.0, "reason": "No new configuration passes the held-out gates; retain standard fallback and inspect audio", "fallback_standard": true }, "reference_only": true, "updated": 1793023747.4814413 }, "rows": [ { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "0034b5d803ff68060564a0e459a5bb711e4444cdd36f997b1459154824eadbe7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "496bf56e7f91d5bde527d45c3a8ff04e67b9ad2ae4d8a8f574280690fecfa2a0", "generation_seconds_batch": 14.904567815014161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7678123712539673, "target_z": 0.8437499336519307, "ce": 5.903473377227783, "genuineness": 4.482243061065674, "identity": 0.528923511505127, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0034b5d803ff68060564a0e459a5bb711e4444cdd36f997b1459154824eadbe7.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "005b4f591780b327a92b5c4ef6104d829938a2b543572e23a98f8ae29b430d98", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "dcfc5c6c2303dbf7756ee8c146c4f568f19cf5e6b4777f140b33cfb8979eb741", "generation_seconds_batch": 20.281186106964014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3695833384990692, "target_z": 0.09716799062958535, "ce": 6.054384231567383, "genuineness": 4.293961524963379, "identity": 0.713862955570221, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/005b4f591780b327a92b5c4ef6104d829938a2b543572e23a98f8ae29b430d98.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "01a3c90078116ad441ee3e1d9e7609856132d13c45591a1d3f6110306a911569", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "cd866a0339c6e6075b2c9262472e6b475f6fefca36ae4c9a26afaec548caf9d2", "generation_seconds_batch": 20.892002947977744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15210701525211334, "target_z": -0.3105468734722484, "ce": 6.078212261199951, "genuineness": 4.545003414154053, "identity": 0.41268572211265564, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/01a3c90078116ad441ee3e1d9e7609856132d13c45591a1d3f6110306a911569.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "01b8a14bb7b2b4e2eea0eb726403d616ebb80410541e5c3bea8e42cc22184b2f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "28612ed35e43d15a85f1d0710b90a043a44b460e8781b9cadfc74e2bc8a9446e", "generation_seconds_batch": 20.557995510054752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017714470624923706, "target_z": -0.5624999930159927, "ce": 6.403861999511719, "genuineness": 4.507347106933594, "identity": 0.570073127746582, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/01b8a14bb7b2b4e2eea0eb726403d616ebb80410541e5c3bea8e42cc22184b2f.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "01ba74c5fdbbe436680eb0817d083e1d57fda127ba790f304c5eaa2e4d6a297e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4d17b42329bd94bf0d07eba5337355f7402644b0debbf59371c581e246d1c3ea", "generation_seconds_batch": 22.400838118046522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19325822591781616, "target_z": -0.23339842800611507, "ce": 5.732706069946289, "genuineness": 4.68307638168335, "identity": 0.6833997964859009, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/01ba74c5fdbbe436680eb0817d083e1d57fda127ba790f304c5eaa2e4d6a297e.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "53c09c3a5168e9b0957ff66322ab94b3de41f9c8ca94ec071b782c2711224a16", "generation_seconds_batch": 15.678969554952346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.0146708488464355, "genuineness": 4.507347106933594, "identity": 0.7477495670318604, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lié le bien, et vérifie les deux derniers chiffres. Appelle-moi, quand tu auras un moment.", "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/emo_Confusion/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "01ffb5af2e6e8bfa6d5fb2795ff786f98202f9503af9e4492f77c59712cef136", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "54403a036c43acc222a84553924090e724be90f50073e907a67f701aa15bc240", "generation_seconds_batch": 16.41328211594373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7219730615615845, "target_z": 0.757812450238948, "ce": 5.9431867599487305, "genuineness": 4.344170093536377, "identity": 0.47624677419662476, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/01ffb5af2e6e8bfa6d5fb2795ff786f98202f9503af9e4492f77c59712cef136.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "028d0d248c30af67d9b3fecf210f991cd0f483328a4044d510cdcd9f528932de", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9886d0b0e30253acd9e375086a21f4d688449acdcc23a078b3fad91e7ed0bd2f", "generation_seconds_batch": 20.52017790300306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023965269327163696, "target_z": -0.5507812757535269, "ce": 6.054384231567383, "genuineness": 4.632867813110352, "identity": 0.6527300477027893, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/028d0d248c30af67d9b3fecf210f991cd0f483328a4044d510cdcd9f528932de.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "029195db929fc25df1b3c2c1b89bf538e4bc4894784f2cf89689013631fec5b4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "38700d6b60b973892e50e9f0de1f59b0670988623762b2cdae0629c9642a332e", "generation_seconds_batch": 21.635982164996676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.443574905395508, "genuineness": 4.6705241203308105, "identity": 0.5014736652374268, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/029195db929fc25df1b3c2c1b89bf538e4bc4894784f2cf89689013631fec5b4.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": "02f875fb61af9a01d3c03cf79278b1568187dcbda1fa59ecc341ec2306dc317d", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8349552ec5ee0bca1f9460909d0d70f5b40dc42fe6690d56fb5b5d70aba40442", "generation_seconds_batch": 21.928311440977268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.259673833847046, "target_z": 3.64062494238194, "ce": 5.673136234283447, "genuineness": 4.444586753845215, "identity": 0.8277712464332581, "hypothesis": "Ton message is arrived after my départ. Dis-moi quand nous pourrons parler.", "wer": 0.5714285714285714, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/02f875fb61af9a01d3c03cf79278b1568187dcbda1fa59ecc341ec2306dc317d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "03e2008e2ebe0ee08b769d2732866b40687c4ccc6b3469ac5237e03ee6371bcf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c00bb5426088f7a2b671e1bd7309dc0e7a611015a84cb8cac7ace426c0918ef3", "generation_seconds_batch": 16.290092927054502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13543817400932312, "target_z": -0.3417968699802448, "ce": 6.117925643920898, "genuineness": 4.65797233581543, "identity": 0.6329348087310791, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/03e2008e2ebe0ee08b769d2732866b40687c4ccc6b3469ac5237e03ee6371bcf.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "03f584ce71f64b7a9de3c5ef810acb89ac4a61a7cf55fd9ecb1d26a9bc7fd150", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9bd65b4c0a2c0262218f6ce4e5bd2cc3c75a063c868815f755a3e14707798532", "generation_seconds_batch": 18.389347480027936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0845204591751099, "target_z": 1.4375000628560657, "ce": 5.303802013397217, "genuineness": 4.2751336097717285, "identity": 0.6151227355003357, "hypothesis": "Die Tür 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/emo_Confusion/03f584ce71f64b7a9de3c5ef810acb89ac4a61a7cf55fd9ecb1d26a9bc7fd150.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "0412145020026d2d1d9ee3b02bed3e7c06b00e2ca668f46a0e367820f72d39a5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c5bc0e0db414369d72a54c6a5e71ff7f2bb203414555b8118f189779d393f7ae", "generation_seconds_batch": 19.320006362977438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1595301628112793, "target_z": 1.5781248934938887, "ce": 5.510311126708984, "genuineness": 4.6705241203308105, "identity": 0.7520380616188049, "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/emo_Confusion/0412145020026d2d1d9ee3b02bed3e7c06b00e2ca668f46a0e367820f72d39a5.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "04449b74de162520f5d01228127ce1c727a016dcf95e46bd6e8c38f44b3fad18", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "15b22a56e0384900e77fb28b815f27f014bebe92388c245138a8b9ae2bb6e457", "generation_seconds_batch": 16.7010849150829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5513784885406494, "target_z": 4.187499755559744, "ce": 5.645336627960205, "genuineness": 4.15588903427124, "identity": 0.11110389232635498, "hypothesis": "De deur staat nog op 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/emo_Confusion/04449b74de162520f5d01228127ce1c727a016dcf95e46bd6e8c38f44b3fad18.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "04792152be2ecf655ec7560ecbf53a7a2dd08a8223286a14d1db9edd1e6d017b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "463ffc25ca214ce5acffe972fbe25572a11a973a54396597e471ea5418ffb586", "generation_seconds_batch": 21.21281185396947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36176979541778564, "target_z": 0.08251951024341556, "ce": 5.704906940460205, "genuineness": 4.557555198669434, "identity": 0.7252859473228455, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/04792152be2ecf655ec7560ecbf53a7a2dd08a8223286a14d1db9edd1e6d017b.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "0488fa5e3c04e6e388816919fc09d753e8cbe5a4616d96d23138e0c4c235255a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5fab04c9f276c749c7e37353cb0259b42d5f5a1e4efe8e70a208e345e4800c6", "generation_seconds_batch": 20.465676228981465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00104561448097229, "target_z": -0.5937500174600182, "ce": 5.895530700683594, "genuineness": 4.256305694580078, "identity": 0.6522963047027588, "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/emo_Confusion/0488fa5e3c04e6e388816919fc09d753e8cbe5a4616d96d23138e0c4c235255a.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "04d0636587dc04bc7c0486caa776b8a19360ea8d4a893bfa9145afdf8d018e54", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4334eeb8e091b279c2c2be5000fe8f56d6e64243954132cc94fb28f3028e624b", "generation_seconds_batch": 20.455518842907622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.288713812828064, "target_z": 1.8203126108711158, "ce": 5.689021587371826, "genuineness": 4.362998008728027, "identity": 0.44004860520362854, "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/emo_Confusion/04d0636587dc04bc7c0486caa776b8a19360ea8d4a893bfa9145afdf8d018e54.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "04fb5872e5de52d41769faf1f3e12a0cfa3e4fe962fbce7f374cbe436264a2fa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0942317f67fac3c3dbeca12828cd3e989454862ddda945a74b67150174b11ae4", "generation_seconds_batch": 15.810208895942196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5552846789360046, "target_z": 0.44531254103104284, "ce": 5.550024509429932, "genuineness": 4.469690799713135, "identity": 0.6425281763076782, "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/emo_Confusion/04fb5872e5de52d41769faf1f3e12a0cfa3e4fe962fbce7f374cbe436264a2fa.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "05966b7eb906fcefeace5b66301f29caa2c68cfc2a67ea0003fdb65493843571", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1a4805b3d2bbd5633b6cd31e5fdb26c84b92defcb8aabee996461c312600a90e", "generation_seconds_batch": 19.320006362977438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9428352117538452, "target_z": 1.171874910953907, "ce": 5.871703147888184, "genuineness": 4.425758361816406, "identity": 0.5518710613250732, "hypothesis": "La puerta sigue 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/emo_Confusion/05966b7eb906fcefeace5b66301f29caa2c68cfc2a67ea0003fdb65493843571.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "30ac9dde6aedf69b78182afee00dcf607ea97d35515522ae45594f09819af424", "generation_seconds_batch": 15.407373536960222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.507116317749023, "genuineness": 4.595211505889893, "identity": 0.6219381093978882, "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/emo_Confusion/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "06049fa5f6c7bbd047b9c68a0c5efa9e9274373c2fea59795085be3a88c376ad", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d9f4d5ba6f55f6c91d948fb53e80f5fcdfc77535fef9ba5cd7fbee37149fd612", "generation_seconds_batch": 15.693516907980666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7365583181381226, "target_z": 0.785156235595485, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.5169193744659424, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/06049fa5f6c7bbd047b9c68a0c5efa9e9274373c2fea59795085be3a88c376ad.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "063eabf914ef56485c4eea89b40039b08403a3a42a9024a6c3c4ff235827b3df", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "851521b5c80bc06f5a5692928db949feb2a35bf874d6c2547ec76c59b6cb24fc", "generation_seconds_batch": 21.201648335088976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3720580339431763, "target_z": 1.9765626213471268, "ce": 6.0146708488464355, "genuineness": 4.419482707977295, "identity": 0.7078878879547119, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/063eabf914ef56485c4eea89b40039b08403a3a42a9024a6c3c4ff235827b3df.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "078c27026806c7eae6e920a542920dfaa3793b22186515b329036f4977e0523d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28e34b38169925e404d15e1815fb0033bd700dc21832b007cb77d12485ab5290", "generation_seconds_batch": 21.634572334005497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11147670447826385, "target_z": -0.38671875916650955, "ce": 5.871703147888184, "genuineness": 4.268857479095459, "identity": 0.4503776431083679, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/078c27026806c7eae6e920a542920dfaa3793b22186515b329036f4977e0523d.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "079b2252506c2b041d36ab94158dcb19bb24219baa9af5bed0691f25c1c41f85", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d4a8fb8924fdf940604bb58581a62abbc665ecd4d9b391dd3ea737c646b6bc4a", "generation_seconds_batch": 21.73185832996387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3695833384990692, "target_z": 0.09716799062958535, "ce": 5.982900142669678, "genuineness": 4.545003414154053, "identity": 0.7140207290649414, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/079b2252506c2b041d36ab94158dcb19bb24219baa9af5bed0691f25c1c41f85.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "07ef83ab6fa70647a2f874d419caa178cbd78aafeb9bd8569d5823aa396addda", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8766b038aea9f634851b5dbfcbc544691c4bf93db4582e698222b213e07256b8", "generation_seconds_batch": 21.605534713948146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06251198053359985, "target_z": -0.47851562914675433, "ce": 5.998785495758057, "genuineness": 4.645420074462891, "identity": 0.611163318157196, "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/emo_Confusion/07ef83ab6fa70647a2f874d419caa178cbd78aafeb9bd8569d5823aa396addda.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "08b2962a4e4c3d50231b83c7b1795530bd1e125800fa4a076541cd16bfbafc36", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "96e6d07eb510ac85653e00cc88cf450e8c43634d1c17f02ecd8bf13613d5bcf0", "generation_seconds_batch": 20.892002947977744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3062938153743744, "target_z": -0.021484370853245668, "ce": 6.109982967376709, "genuineness": 4.821148872375488, "identity": 0.5500320792198181, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/08b2962a4e4c3d50231b83c7b1795530bd1e125800fa4a076541cd16bfbafc36.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "08e8d3e675169ee80ac43c88a24839377b6f2537da9ec968496805d41c9f789f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8380b8a73252c11f8b34375b7955034d01c7f07930437df36c048661a61a8482", "generation_seconds_batch": 19.380240180995315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34171509742736816, "target_z": 0.04492186125023564, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.23316466808319092, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/08e8d3e675169ee80ac43c88a24839377b6f2537da9ec968496805d41c9f789f.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "0942659c899de6562285f7d81cb12949308b12d0ced31cf7cbb2d45aae37fd6a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "148e425de4e88bd9bad254f87642711bc9f07218ece6aa5dafec93e1449799a0", "generation_seconds_batch": 20.450848050997593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2180018424987793, "target_z": 3.5625001606321676, "ce": 5.895530700683594, "genuineness": 4.13078498840332, "identity": 0.29263150691986084, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0942659c899de6562285f7d81cb12949308b12d0ced31cf7cbb2d45aae37fd6a.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "0a098b5543836008cc79f45d43f4699ca349c5ff1b856d6a0b3410e4eb86e071", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3c1964312ca4a84c5a7759cb55032f973739e874a3d7cd8bb6aee373d709fc7a", "generation_seconds_batch": 22.686378529062495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7262709140777588, "target_z": 2.6406250541260565, "ce": 5.2343034744262695, "genuineness": 4.369274139404297, "identity": 0.5172470808029175, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0a098b5543836008cc79f45d43f4699ca349c5ff1b856d6a0b3410e4eb86e071.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "0a73d46f35c6514c296db4c1a1c664e0ad8568f240b0e2b99200e8d024cd405a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1420d71dd107e1bd3fdc7522c9ea9a13fdc6c9bd0fd943f9171e883180ea844f", "generation_seconds_batch": 21.047677561058663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.5657335519790649, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0a73d46f35c6514c296db4c1a1c664e0ad8568f240b0e2b99200e8d024cd405a.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "0aa3e112001bba135e5cf08921f9805463abe82fd349c28dbff4d33130d00aaa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ce707d90ff97ecdcea91c938ea7c6a3c0480c30189e6d452c362f572363d20ab", "generation_seconds_batch": 25.231844969908707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.09409761428833, "genuineness": 4.65797233581543, "identity": 0.8176467418670654, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0aa3e112001bba135e5cf08921f9805463abe82fd349c28dbff4d33130d00aaa.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "0abb5540a98a72f9629bbc623f5043d3ccfdccd126d0b1aa50d3ba75950ade70", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "34fe4b67800a6c76592ad7b5651f59e416b14c1c97cb76e690128be38103fa6a", "generation_seconds_batch": 16.134654072928242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2930108308792114, "target_z": -0.04638671487605846, "ce": 6.078212261199951, "genuineness": 4.6705241203308105, "identity": 0.5942853689193726, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0abb5540a98a72f9629bbc623f5043d3ccfdccd126d0b1aa50d3ba75950ade70.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "0b1cd6cda5cfc6658986c5c32dab713945d0cd2161847153daffff0e443ef887", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "036056042d43f0e51aa4638985bda7b0ee718f9c08e4f494f64bffcd319b5c36", "generation_seconds_batch": 16.22359822492581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07605542242527008, "target_z": -0.4531249912699909, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.44207918643951416, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0b1cd6cda5cfc6658986c5c32dab713945d0cd2161847153daffff0e443ef887.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": "0b225c18213d46666891a792997a87cee5ddb2d66b44ff841697dfaff9751eb4", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fde67b479b64bf21b11d2837b21016299df914f6c9bd7876ef0a6d9f4f9b695", "generation_seconds_batch": 15.498714011977427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.301215410232544, "target_z": 1.8437500453960474, "ce": 5.685050010681152, "genuineness": 4.256305694580078, "identity": 0.43250733613967896, "hypothesis": "The keys are n the kitchened table. I checked the drawer, then looked beside the window. Take them with you, Anuline.", "wer": 0.21739130434782608, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0b225c18213d46666891a792997a87cee5ddb2d66b44ff841697dfaff9751eb4.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "0baad3a2783de38deff787413e5d4ff3c5181421273c11173a91620306b60a11", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5f7517285c1a692a445b57d21cead927d247fdec894cdfd67eb207fd7a88c339", "generation_seconds_batch": 13.717525087064132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5927895307540894, "target_z": 0.5156249563499544, "ce": 6.038498878479004, "genuineness": 4.362998008728027, "identity": 0.6667599081993103, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0baad3a2783de38deff787413e5d4ff3c5181421273c11173a91620306b60a11.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": "0bbb1be989ac9d1a72d59d980465dc2b5115e5bed31b95b3508623e337137505", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb85308310b6dd1575ec9e4767d3c8ea343357dbc74c6d7875a71c674b82c18b", "generation_seconds_batch": 20.784826064947993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7929463386535645, "target_z": 2.765625151902159, "ce": 5.526196479797363, "genuineness": 4.193545341491699, "identity": 0.5178260803222656, "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/emo_Confusion/0bbb1be989ac9d1a72d59d980465dc2b5115e5bed31b95b3508623e337137505.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "0c1a547e2efb84f6009f821d64fdfeccc349fe047c0c56e66cd6eab9ef5e0027", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 131, "seconds_generated": 10.48, "mp3_sha256": "b01a880c88346017493a3ec682c3447a3ee17a0cf80f0ad375ced8a7eab5f274", "generation_seconds_batch": 15.590884095989168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.213703989982605, "target_z": 1.6796875567450593, "ce": 5.824047088623047, "genuineness": 4.369274139404297, "identity": 0.4441465735435486, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du beast.", "wer": 0.10526315789473684, "duration_error_s": 0.28000000000000114, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0c1a547e2efb84f6009f821d64fdfeccc349fe047c0c56e66cd6eab9ef5e0027.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "e865fcc6e7cc57848edc0febc623d4bc9e4486600d8ac2e23c803a3985a676ca", "generation_seconds_batch": 15.188428443972953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0698046088218689, "target_z": -0.46484373646848587, "ce": 6.252951145172119, "genuineness": 4.231201171875, "identity": 0.8717504739761353, "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/emo_Confusion/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "0d1a0123d87d11a06f1a6c7d3e016fb10eae4c87a2c5b689e75a3d134c0e4150", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fe85be0439651087632a0d7a6dffe40e03cdbe511849c772e69c43718469099b", "generation_seconds_batch": 20.5009416899411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07605542242527008, "target_z": -0.4531249912699909, "ce": 6.23706579208374, "genuineness": 4.7834930419921875, "identity": 0.6260663270950317, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0d1a0123d87d11a06f1a6c7d3e016fb10eae4c87a2c5b689e75a3d134c0e4150.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": "0d84eb015f3d1ce58a5b5f7cea77f00239ecd003a042c0ef21498027d8c58991", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbee27d0c8ff8502ff5f8751ee4cce033e5a3c2bde1465332a7fa8074f25822f", "generation_seconds_batch": 25.94291638804134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.363723635673523, "target_z": 1.9609376649971724, "ce": 5.708878040313721, "genuineness": 4.2249250411987305, "identity": 0.28741830587387085, "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/emo_Confusion/0d84eb015f3d1ce58a5b5f7cea77f00239ecd003a042c0ef21498027d8c58991.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "0dd288e1abe7d5e7c8650d9101a7eadddee4ac91ece441d61b05291d1a915eee", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b586e91d4f87897f68488add6ee13df02bb87f57f6538c297eaea81ae1c815bc", "generation_seconds_batch": 21.242282197927125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9720057249069214, "target_z": 1.226562481666981, "ce": 5.649308204650879, "genuineness": 4.607763767242432, "identity": 0.5185489654541016, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0dd288e1abe7d5e7c8650d9101a7eadddee4ac91ece441d61b05291d1a915eee.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "0df97717bfe4a3fdf0c1404500e5d4b19033b94c82a7f5739406e65fa4fb0dfa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b27348a7cbf9913c7571cbc2fbee7e61609dfe62d8fad3281b90a9bb78cc647a", "generation_seconds_batch": 23.959280307986774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6740501523017883, "target_z": 0.6679687277384767, "ce": 5.982900142669678, "genuineness": 4.545003414154053, "identity": 0.6203292012214661, "hypothesis": "Die Tür 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/emo_Confusion/0df97717bfe4a3fdf0c1404500e5d4b19033b94c82a7f5739406e65fa4fb0dfa.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "0e1e5137df4c6bb1acbf668530eec55467bc83dab65ba34bc1455e5eca7e807c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b6acc9fabb1901a6d8e238c12f767ffb84b72702252f37b1875774d7ffd8fa78", "generation_seconds_batch": 19.58181109803263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2312840223312378, "target_z": -0.16210936910724386, "ce": 6.09409761428833, "genuineness": 4.595211505889893, "identity": 0.5181261301040649, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0e1e5137df4c6bb1acbf668530eec55467bc83dab65ba34bc1455e5eca7e807c.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "0e591315dd2c8715edc6387649137dfa97a0183e736188bf04c3f13114715451", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "57f37b109ae9f47bb46880c8fcb389cda1548edc716507036cf36e4f83dc0594", "generation_seconds_batch": 15.398985911044292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4679038524627686, "target_z": 2.1562500663480693, "ce": 5.704906940460205, "genuineness": 4.532451152801514, "identity": 0.31768500804901123, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0e591315dd2c8715edc6387649137dfa97a0183e736188bf04c3f13114715451.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": "0e5f3ed43641ce62a8b550e3bca5ad5442403797a823747131d56e02fa233861", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8012314ff8111d031dd7a3f67cddbd4edf9f90e68f3977817172125ad3c5dd0", "generation_seconds_batch": 15.314731306047179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8886615037918091, "target_z": 1.07031247119097, "ce": 5.669164657592773, "genuineness": 4.582659721374512, "identity": 0.7895131707191467, "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/emo_Confusion/0e5f3ed43641ce62a8b550e3bca5ad5442403797a823747131d56e02fa233861.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "0eb4a56e724beaa492cb4ba283ff91209a314b5e97682078bb0944259be7311a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e14dbbcb5767b691a714e59932fecc49b964ca49d81115471c466003b4e81fc5", "generation_seconds_batch": 20.281186106964014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9178320169448853, "target_z": 1.1250000419040438, "ce": 6.348263263702393, "genuineness": 4.632867813110352, "identity": 0.6212873458862305, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/0eb4a56e724beaa492cb4ba283ff91209a314b5e97682078bb0944259be7311a.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "10101e6fd799af613c8baae151c559b48fc5396122ed95e7bcf1544846c50b11", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d23067a646230f00b563f16c6c159e5ab1dcc00651d9c66fb1dc9558a575446f", "generation_seconds_batch": 20.808632764965296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.280379295349121, "target_z": 1.804687431032928, "ce": 5.645336627960205, "genuineness": 4.394378662109375, "identity": 0.6758944988250732, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/10101e6fd799af613c8baae151c559b48fc5396122ed95e7bcf1544846c50b11.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "1033a4d79e631d4af95e85178212d2a166cb7c5b3f5cee018fd77b4ac1b4e66c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de4da3611e4321feb4491af4069f68473addbb486fc1d88eea77d1303dd2731c", "generation_seconds_batch": 22.024176714010537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47506582736968994, "target_z": 0.29492183331420646, "ce": 5.720792293548584, "genuineness": 4.149612903594971, "identity": 0.3023270070552826, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1033a4d79e631d4af95e85178212d2a166cb7c5b3f5cee018fd77b4ac1b4e66c.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": "105ae982e3e4f43915d7acdf654aeec4c7048ccaa90495eb47b091118959ebea", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c39ba945df6510b52176562215dba3f7b50b6662b645c6374d77fe8350c65f1", "generation_seconds_batch": 21.344913888024166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.768073558807373, "target_z": 4.59374996158796, "ce": 5.3792572021484375, "genuineness": 4.268857479095459, "identity": 0.31435704231262207, "hypothesis": "De sleutel liggen op de keukentafel. Ik kijk in de la en 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/emo_Confusion/105ae982e3e4f43915d7acdf654aeec4c7048ccaa90495eb47b091118959ebea.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": "1063ca9630e23639a9ce0cd76ec4e03cafe99ffdf3089ef386862f826fc9da72", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e63ed560ab5b5dc88883c43858b639fba1cb582dccbcbe945be0e79008f3773", "generation_seconds_batch": 15.498714011977427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2849368453025818, "target_z": -0.06152346118014973, "ce": 5.796247482299805, "genuineness": 4.1998209953308105, "identity": 0.32186686992645264, "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/emo_Confusion/1063ca9630e23639a9ce0cd76ec4e03cafe99ffdf3089ef386862f826fc9da72.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "111f71f9d4963b8cc4740a07af8021d0ccbe19d90185dd865d4f236376b5718c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7de098c267a4669e81560ed2980b0b7975eeb2b6b44b019ae108f9ecbc7432ed", "generation_seconds_batch": 16.17279724299442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7512742280960083, "target_z": 2.6875001466641533, "ce": 5.5738525390625, "genuineness": 4.218649387359619, "identity": 0.4787059426307678, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/111f71f9d4963b8cc4740a07af8021d0ccbe19d90185dd865d4f236376b5718c.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": "11a3c4d3433c0343fd8ac0e5427c7a82055211ef4e001d1039cdad8a85ec28f3", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3200249b243e129d21a4b25ddb256b84a977657b31f2fba0d9699a535af0648c", "generation_seconds_batch": 24.104134701075964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10939310491085052, "target_z": -0.39062499825399816, "ce": 6.141753673553467, "genuineness": 4.65797233581543, "identity": 0.3122677206993103, "hypothesis": "I wrote the number on the envelope, read it carefully, and checked the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/11a3c4d3433c0343fd8ac0e5427c7a82055211ef4e001d1039cdad8a85ec28f3.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "11d040a361d2ab2be33b765a5e34aebf564b2c7a845551006d0035a3256dcce9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2fa53334c384a7789b34b1ee5a9730b41ba67a7c50af34a0c0535fc99c0c0db7", "generation_seconds_batch": 21.766129560070112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8303205966949463, "target_z": 0.9609375532530556, "ce": 5.7684478759765625, "genuineness": 4.65797233581543, "identity": 0.6499292850494385, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/11d040a361d2ab2be33b765a5e34aebf564b2c7a845551006d0035a3256dcce9.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "1216c90eb6942c12c53a4aee75284760877bfc9fe4865aab5534ef076e33439d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a8a4989ef0f170c6faa2425d79a3b45848041376546377a86db833fdf0c541f", "generation_seconds_batch": 15.628392399055883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3422360122203827, "target_z": 0.04589844895813697, "ce": 5.919358730316162, "genuineness": 4.883909225463867, "identity": 0.7685558795928955, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1216c90eb6942c12c53a4aee75284760877bfc9fe4865aab5534ef076e33439d.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc05c98ec02b0f137488abfc06cc704671bffd23481025c610dcea4f99663b63", "generation_seconds_batch": 15.384951777989045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.011456012725830078, "target_z": -0.6171875078570082, "ce": 6.197352409362793, "genuineness": 4.369274139404297, "identity": 0.7507076263427734, "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/emo_Confusion/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "12c069fdbd4e83db85b3ee6aae6da3ad2fa35cf2358cd5f532f292bc4283c407", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "744db6eacf2dbea2f469829e647e7f31c902975274f96631430d8d046f40826f", "generation_seconds_batch": 19.44128825608641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35161224007606506, "target_z": 0.06347658072389403, "ce": 6.038498878479004, "genuineness": 4.545003414154053, "identity": 0.5012667179107666, "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/emo_Confusion/12c069fdbd4e83db85b3ee6aae6da3ad2fa35cf2358cd5f532f292bc4283c407.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "12c6672700f01c5c27d06c76ede6e22f42576d06ccd165f5722982ee03a0250e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8567edd02ec807ac87863cad459c26556aa12e662ba7b8869e343797b1c33444", "generation_seconds_batch": 22.397199576022103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06563737988471985, "target_z": -0.47265627051552145, "ce": 6.133810997009277, "genuineness": 3.835810661315918, "identity": 0.3424600660800934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/12c6672700f01c5c27d06c76ede6e22f42576d06ccd165f5722982ee03a0250e.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "12df566d81234b3f420f97b4ff2244b99fd481bf5ab63f79de201ef5d1f90697", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ca747b5c447417ffc5fa750a330701c6fecc10a36c213c0073d0998d1bef3685", "generation_seconds_batch": 14.931639419985004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.005212843418121338, "target_z": -0.5859374834129827, "ce": 6.356205940246582, "genuineness": 4.457139015197754, "identity": 0.5437277555465698, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/12df566d81234b3f420f97b4ff2244b99fd481bf5ab63f79de201ef5d1f90697.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": "130222aa2308c2dafcb6d5516162632ff4a14d3e965181563cddca5eca79a630", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a87af31b0a28b8a59c299c0b4368d48bffd70b8bff9cec670452f5f4fbd77f94", "generation_seconds_batch": 13.613664992968552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1763296127319336, "target_z": 3.4843749319059287, "ce": 5.597680568695068, "genuineness": 4.1998209953308105, "identity": 0.31141576170921326, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen.", "wer": 0.043478260869565216, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/130222aa2308c2dafcb6d5516162632ff4a14d3e965181563cddca5eca79a630.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "136a671f5557c2476cb5f14b50f34351b172147ff4abecc50c830e1a2b404700", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "93ce7ff544e71fe77625af2733a511be191f056f70d76a97cf2cbd363c67f3cd", "generation_seconds_batch": 16.7010849150829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9596346616744995, "target_z": 3.078124949365947, "ce": 5.855817794799805, "genuineness": 4.162164688110352, "identity": 0.44093284010887146, "hypothesis": "De deur staat nog open. Ik heb je het 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/emo_Confusion/136a671f5557c2476cb5f14b50f34351b172147ff4abecc50c830e1a2b404700.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "139b43b0c8e4c54e3d36ed6363e23c4675ec74f7b31d07c3f3b11b56ab2cd428", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17bafbe153f16254e817dfc57429fad2ded30fa52be328c1fec8919d9f2d2b5b", "generation_seconds_batch": 15.923169014975429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23909753561019897, "target_z": -0.14746094459313241, "ce": 6.23706579208374, "genuineness": 4.645420074462891, "identity": 0.359027624130249, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/139b43b0c8e4c54e3d36ed6363e23c4675ec74f7b31d07c3f3b11b56ab2cd428.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "13b8f78d35815a37f10e3ffa550bee7613afcfe4c13b12d1dce2061e92912dfc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "47de8d188e675bed23380aa4c70992f369dbe05b9c86170b6805ae1ad82c3e23", "generation_seconds_batch": 20.67374404403381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846118211746216, "target_z": 2.749999972063971, "ce": 6.038498878479004, "genuineness": 4.557555198669434, "identity": 0.6019361019134521, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/13b8f78d35815a37f10e3ffa550bee7613afcfe4c13b12d1dce2061e92912dfc.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "13c4d9930c5270726426237b69ad03528bd2676f3574e645a185cd7f8edff689", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2b2303915da52153b0246732ffe906818059bf7983f225e680f505f4a58dcfb9", "generation_seconds_batch": 21.208438965957612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.133810997009277, "genuineness": 4.632867813110352, "identity": 0.5523690581321716, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/13c4d9930c5270726426237b69ad03528bd2676f3574e645a185cd7f8edff689.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "1467a8cd761f67754be00da58e8d8ae7cc31c388cdade33c4466fa6572cef542", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0ceb7d13f7282ae680b789a6ba20a1dd8c2836093deaf6942f8d621a2fb487bd", "generation_seconds_batch": 15.693516907980666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32660895586013794, "target_z": 0.01660154405785574, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.3630191683769226, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1467a8cd761f67754be00da58e8d8ae7cc31c388cdade33c4466fa6572cef542.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "14953bc0906cd908d61c1415c6657dda50f243c93b15bf9d8f75580593412507", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f152339e2ba4b6fa5780101d0a7cd72cf33fc2a65855c6871b5d2b0e10b5de5", "generation_seconds_batch": 20.509842209983617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11981113255023956, "target_z": -0.3710937469444968, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.3677004873752594, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/14953bc0906cd908d61c1415c6657dda50f243c93b15bf9d8f75580593412507.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "14ac24bd1e6c9eaa8f549454af48f17c39249a9653ee90788e48978038f9cbf2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e6648701a87ae471c518f400c31dcafaf94915ba00d9375b569b9d4665512198", "generation_seconds_batch": 14.893556658993475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9386680126190186, "target_z": 1.1640624327789297, "ce": 5.935244083404541, "genuineness": 4.3504462242126465, "identity": 0.19793826341629028, "hypothesis": "I found the little box. It was beside the old window we can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/14ac24bd1e6c9eaa8f549454af48f17c39249a9653ee90788e48978038f9cbf2.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "1552a8d37652f1c10b6f2306597689c2a2c7f88bbda8c758f333a8d2c016f124", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "83ae950572dd21632c5a60501b1a036886097ddec46e49de4dc4eaea11d1ffd1", "generation_seconds_batch": 19.666661383002065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0886876583099365, "target_z": 1.4453125410310428, "ce": 5.34748649597168, "genuineness": 4.243753433227539, "identity": 0.6556786298751831, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1552a8d37652f1c10b6f2306597689c2a2c7f88bbda8c758f333a8d2c016f124.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "85ad93ed17d25881ed8edd0694821cffb50591cca7883a1a4e00d1f22a3a09bb", "generation_seconds_batch": 14.294577624998055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.011463642120361328, "target_z": -0.5742187661505168, "ce": 6.292664527893066, "genuineness": 4.306513786315918, "identity": 0.3321572542190552, "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/emo_Confusion/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "15aa54863230296ad4d44a395e169153546a4f2f926aec80881839779e3d4dee", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "72104fabb8edd5f43d6aa7671a28aad65d87c11b6f58f6cbb868336089c1930b", "generation_seconds_batch": 15.398985911044292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8761599063873291, "target_z": 1.0468750366660382, "ce": 6.070269584655762, "genuineness": 4.482243061065674, "identity": 0.5232356786727905, "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/emo_Confusion/15aa54863230296ad4d44a395e169153546a4f2f926aec80881839779e3d4dee.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "15d4ae448f32258f41fed333598f21a29d0363e102ed3f5006d2557db1e0a9c7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "14900ed88d4185e2cce75f892e7be0448239a624de0c84464c919329358510fb", "generation_seconds_batch": 23.6033853800036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7596086263656616, "target_z": 2.7031251030141075, "ce": 5.510311126708984, "genuineness": 4.582659721374512, "identity": 0.7051901817321777, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/15d4ae448f32258f41fed333598f21a29d0363e102ed3f5006d2557db1e0a9c7.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": "168f25270b39b97dd8713a5ab73f35bf404e3c450ce182c1681aab8fbf2d291c", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d901a6bb18f1353e66cb0b61ae377da2063ac6e898bf786be2b009f638dbf18b", "generation_seconds_batch": 16.22802845109254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6865518093109131, "target_z": 0.6914062740075251, "ce": 5.919358730316162, "genuineness": 4.13078498840332, "identity": 0.5118935108184814, "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/emo_Confusion/168f25270b39b97dd8713a5ab73f35bf404e3c450ce182c1681aab8fbf2d291c.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "16b2c12c25f086c4b7802b4f6148b3c411129f2d8d4141ae5bd60a61e4885b5b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8f29f2d009304b3e6922d4c3aa944d232ba843f51277b160bd48f1e80d234a79", "generation_seconds_batch": 15.944600107031874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7448927164077759, "target_z": 0.8007811919454394, "ce": 5.661221981048584, "genuineness": 4.7834930419921875, "identity": 0.8304545879364014, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/16b2c12c25f086c4b7802b4f6148b3c411129f2d8d4141ae5bd60a61e4885b5b.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": "16f172448c23d134f3986c23e70157531cca5b1b619476fd138928828042ff89", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ff2fa138238c250ac600e1b192156addc138188f3ea026f1f5e48ca84203903", "generation_seconds_batch": 16.24333581002429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.767943024635315, "target_z": 2.718750059364062, "ce": 5.426913261413574, "genuineness": 4.400654315948486, "identity": 0.5487565994262695, "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/emo_Confusion/16f172448c23d134f3986c23e70157531cca5b1b619476fd138928828042ff89.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "16fe51a5b48009cb2a739eab219df525de9ff312419e0ede1ac7b9bc3eb793f3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ce0f4b64be9830b1d0f27ac1bc8e18441f0b9a2c3ce6ae32bcb2cd9707aa0949", "generation_seconds_batch": 21.750097501906566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7512742280960083, "target_z": 2.6875001466641533, "ce": 6.054384231567383, "genuineness": 4.482243061065674, "identity": 0.5106561779975891, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/16fe51a5b48009cb2a739eab219df525de9ff312419e0ede1ac7b9bc3eb793f3.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "17000afacb3d0ef2f99bdb0bf1987a516a7ef344ae358e725852f169a314f49f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "78e923e9ebabac4690b25000c948ef4e0e53236d321c517aca28ae502f1170fc", "generation_seconds_batch": 16.000706993043423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 5.692992687225342, "genuineness": 4.557555198669434, "identity": 0.7164442539215088, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/17000afacb3d0ef2f99bdb0bf1987a516a7ef344ae358e725852f169a314f49f.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "171eb84c23f8f0e4fe9fd154fb2b5751d4b4b99f3050d8977401e9c51e3a773f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "27b3487d49c08c3b4e28fc4f7622a35d5970678a30efcad0d6938698a82ae192", "generation_seconds_batch": 21.766129560070112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23857663571834564, "target_z": -0.14843750436500455, "ce": 5.696964263916016, "genuineness": 4.607763767242432, "identity": 0.6613145470619202, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/171eb84c23f8f0e4fe9fd154fb2b5751d4b4b99f3050d8977401e9c51e3a773f.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": "1742679dc0002e75ee7d1a9b2a5344266cc58a86af8d287e0ea8c95ce1a2cb55", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11ddf0c54a9fa659be88c9fd6cff7404d2099f0bdc677453cf77fd05c84a2f10", "generation_seconds_batch": 16.275952197960578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7344747185707092, "target_z": 0.7812499965079963, "ce": 5.760505676269531, "genuineness": 4.281409740447998, "identity": 0.8034316301345825, "hypothesis": "La piste ma route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1742679dc0002e75ee7d1a9b2a5344266cc58a86af8d287e0ea8c95ce1a2cb55.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "175ffcf5a271648b8736694854c89a005d1b7587adbed2b217e0945e05341bc7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7395844c98aabe99167cff211afdd6509abe934478a61fa91393fbac46b2611d", "generation_seconds_batch": 20.703628365998156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7698960304260254, "target_z": 0.847656284483536, "ce": 5.77639102935791, "genuineness": 4.469690799713135, "identity": 0.46861207485198975, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/175ffcf5a271648b8736694854c89a005d1b7587adbed2b217e0945e05341bc7.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "178443957a85d7b0c8b4f101c687a7e0cbaa62c90d06afbd084ff3798ce182ff", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a127eb3ae8314f9943c3716dd33e96170924a66177d74651e347f07e67e66ae5", "generation_seconds_batch": 20.145719173015095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.165581703186035, "genuineness": 4.871357440948486, "identity": 0.7231601476669312, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/178443957a85d7b0c8b4f101c687a7e0cbaa62c90d06afbd084ff3798ce182ff.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "187e9b3c2dfcfeda95d17ad6f51609ca0e4f418ab2229ead1e1494b905d58cda", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2826b1beff67b54880bab9d41bda938e908df2e71a700287b69387031eaf2dc4", "generation_seconds_batch": 15.989142192993313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33858969807624817, "target_z": 0.03906250261900274, "ce": 6.38797664642334, "genuineness": 4.632867813110352, "identity": 0.594886302947998, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/187e9b3c2dfcfeda95d17ad6f51609ca0e4f418ab2229ead1e1494b905d58cda.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "188754d570888924a9775debf65a8649a1817b2c03cbceb1b800fd7b06cc19b0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "bfd672beaca70ab7978e8d67b1154d3a8e564312d0e9613f50760307aa39990b", "generation_seconds_batch": 20.315113247954287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07813902199268341, "target_z": -0.4492187521825023, "ce": 5.95112943649292, "genuineness": 4.519899368286133, "identity": 0.5115861296653748, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/188754d570888924a9775debf65a8649a1817b2c03cbceb1b800fd7b06cc19b0.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "18dddf32f5564f42e49c1d2abe3af7bcbebda85abac721e588262e25042e2c69", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e26fb98609773121d75eb0519202777824b25453500867837ff44f38e1c1fa7f", "generation_seconds_batch": 14.931731909979135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15419061481952667, "target_z": -0.3066406343847598, "ce": 6.284721851348877, "genuineness": 4.381826400756836, "identity": 0.5734243988990784, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/18dddf32f5564f42e49c1d2abe3af7bcbebda85abac721e588262e25042e2c69.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "19c800c33cdb1a34762f4107b80de9d89ff1e3f755a89c86e3ece3761cd40128", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "da6d4bfb71d06f2e3caed81107c526d7a11b5a9e8d9d6fea85f07f31a6acd650", "generation_seconds_batch": 16.134654072928242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0812644362449646, "target_z": -0.4433593656152402, "ce": 5.77639102935791, "genuineness": 4.833701133728027, "identity": 0.7828339338302612, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/19c800c33cdb1a34762f4107b80de9d89ff1e3f755a89c86e3ece3761cd40128.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": "1a0083f69e9e215a9caffc35c905dc486d878e95770352d4ce00a403a93fad71", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7594552f019a722ae23716df79e92e5bfaf05b0e2d98db19b9ca73c1f004ef0f", "generation_seconds_batch": 20.381754116970114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.842952847480774, "target_z": 2.8593751134901186, "ce": 5.450741291046143, "genuineness": 4.494794845581055, "identity": 0.6194392442703247, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1a0083f69e9e215a9caffc35c905dc486d878e95770352d4ce00a403a93fad71.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "1a246c94bc966339b47e3af496b637b51a06c41fff597c66d80ee62653f34c10", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "71e4096c2d72dd65c4235c3534239332e7cca3bbde565525179c8ccf7515aa45", "generation_seconds_batch": 19.380240180995315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6969698667526245, "target_z": 0.7109375811890848, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.6086400747299194, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1a246c94bc966339b47e3af496b637b51a06c41fff597c66d80ee62653f34c10.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "1a697a17683f6b9df367bac7cd3232d5164e8db411c8464b7ed104cca8764b66", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c8e6d075ab2f38ffb789b49263b858f5fa1f61124e5105bd4ab0a9e9bfd69959", "generation_seconds_batch": 16.170916605042294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7929463386535645, "target_z": 2.765625151902159, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.5939935445785522, "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/emo_Confusion/1a697a17683f6b9df367bac7cd3232d5164e8db411c8464b7ed104cca8764b66.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "1a9e599d8f61b1e2760436ea8801f2a553ae4e9d3f1bc79e221beede222ac339", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec6bf49cbbde511eebc73221c4381dac1a3bb76223b60e5ef2fb841d801383a6", "generation_seconds_batch": 20.465764509979635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 6.308549880981445, "genuineness": 4.444586753845215, "identity": 0.48740220069885254, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1a9e599d8f61b1e2760436ea8801f2a553ae4e9d3f1bc79e221beede222ac339.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "1ad68d0588f0f25fc5b2dc3541609ee8d42c508c927510fd187286d488b2b74d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6e21339b504bd49ec56e27eb46a0a2f311ddff734ef05b8752decd826e2f1ea6", "generation_seconds_batch": 26.24704099202063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7553107738494873, "target_z": 0.8203124991269991, "ce": 5.689021587371826, "genuineness": 4.507347106933594, "identity": 0.6842223405838013, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1ad68d0588f0f25fc5b2dc3541609ee8d42c508c927510fd187286d488b2b74d.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "1b37ce45ff1fe8d01541e92ed60fb83856c62ef524964d3080b842f654a06b27", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d0937c3884479364198d271de982a4ed774a8a49f2e7e61de4faf9c6af132777", "generation_seconds_batch": 15.820332654053345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4013588428497314, "target_z": 3.906249647307632, "ce": 5.055593490600586, "genuineness": 3.9425034523010254, "identity": 0.08310499042272568, "hypothesis": "De er staat nog open. Ik heb je juist op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1b37ce45ff1fe8d01541e92ed60fb83856c62ef524964d3080b842f654a06b27.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "1b6426ec822589025e4da90e78e4be834aa9350233ef51e57a37ffa871f13c37", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "19b0b57108f91d99ddd8c561be242ca40ab3e3bad7141b182c37cd8121834a96", "generation_seconds_batch": 20.41972928098403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017714470624923706, "target_z": -0.5624999930159927, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.7616361379623413, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1b6426ec822589025e4da90e78e4be834aa9350233ef51e57a37ffa871f13c37.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "1bb287e9ad576b1025d9f89d6911decc9c22e07783d8c89beb078473d478bcb0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0c3c9185a6b629940ea3ac236badb237df1d2d9d12f40afaf8862841bd56a852", "generation_seconds_batch": 21.177059186040424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.005212843418121338, "target_z": -0.5859374834129827, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.6074201464653015, "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/emo_Confusion/1bb287e9ad576b1025d9f89d6911decc9c22e07783d8c89beb078473d478bcb0.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "1c745c88a4ce891cfb955f29aa169b822c5e2a08d77f95efae80b2a334ca1082", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "938466718e031c9d8e09d61f4f7df6ea2258d3b0366c26935657835b3929980d", "generation_seconds_batch": 21.136447174008936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.8043632507324219, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1c745c88a4ce891cfb955f29aa169b822c5e2a08d77f95efae80b2a334ca1082.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "1c7bba64495af0a0233366b39ac922ee92da9c7c3e83a30bfef96c7237a86f44", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7071e78b738ec919169cd42464f7055471ef86abe03e8f2c5ade0f2a424c9ced", "generation_seconds_batch": 19.162064046016894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1845334768295288, "target_z": 1.6249999860319855, "ce": 5.903473377227783, "genuineness": 4.632867813110352, "identity": 0.5498459339141846, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1c7bba64495af0a0233366b39ac922ee92da9c7c3e83a30bfef96c7237a86f44.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": "1d48b79f8b8abe2e73a4ddddcba77192a2c6c8978112a74540494f5d32b57097", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "218850b23abd257359140cf4d30b740ca9825d81327584b4fe48f05b6f911e4b", "generation_seconds_batch": 20.067378230043687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4464162588119507, "target_z": 0.24121090618109228, "ce": 5.863759994506836, "genuineness": 4.13078498840332, "identity": 0.32450681924819946, "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/emo_Confusion/1d48b79f8b8abe2e73a4ddddcba77192a2c6c8978112a74540494f5d32b57097.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "1d5074e23b4b66a7100f743998f9f1089cb28e74ece2d677c77ceaaf153587a3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ac958ae43c18be1738ea22ce7c83519154ad558d0768fb40a7b88b8a3f2848a3", "generation_seconds_batch": 20.218488796963356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.134657621383667, "target_z": 3.406250150156157, "ce": 5.692992687225342, "genuineness": 4.061748504638672, "identity": 0.6291983723640442, "hypothesis": "De deur 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/emo_Confusion/1d5074e23b4b66a7100f743998f9f1089cb28e74ece2d677c77ceaaf153587a3.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "b5187c08de676314772b968634118ab6c1d319801468fbf0004815b9a5821115", "generation_seconds_batch": 17.747103595989756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.292664527893066, "genuineness": 4.256305694580078, "identity": 0.6228851079940796, "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/emo_Confusion/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "6471b5e3dc3fc40fc7e31c9300dbb2d67440fab8764ec7eafda739fb6e8f22c8", "generation_seconds_batch": 16.30192103504669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10105867683887482, "target_z": -0.40625001047601095, "ce": 6.308549880981445, "genuineness": 4.482243061065674, "identity": 0.36892616748809814, "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/emo_Confusion/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "1e0f78deee571e67cbf28369493c2ab2bdae1c55721d9a42332250ce9773a632", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "52475acc91bd7588ab5f5e56c63ea241adec2baea82af05c6a71fe12dda954d8", "generation_seconds_batch": 19.589918283978477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1073095053434372, "target_z": -0.39453123734148676, "ce": 5.95112943649292, "genuineness": 4.770940780639648, "identity": 0.7379159927368164, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1e0f78deee571e67cbf28369493c2ab2bdae1c55721d9a42332250ce9773a632.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "3bbab5a6ddba46c60e7b94619dd5811ab85bace408a6bd190c3be5ef0eff3715", "generation_seconds_batch": 17.747103595989756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.056261152029037476, "target_z": -0.4902344022812785, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.5424128770828247, "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/emo_Confusion/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.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": "1ec6392bdfc67ee1e35690016e08780ad3d90b3d4784b56fb37d29cd15751d1d", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6d01e3a9b876f926207e5290af9d175322e176cd50c2065ba7bbb1fc85600f0", "generation_seconds_batch": 17.23186749301385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9012937545776367, "target_z": 2.9687500314280326, "ce": 5.696964263916016, "genuineness": 4.394378662109375, "identity": 0.48118603229522705, "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/emo_Confusion/1ec6392bdfc67ee1e35690016e08780ad3d90b3d4784b56fb37d29cd15751d1d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "1f3746761533cd6d7264cd9a3db7d4aca8a132aa1b0c1a206bafdd3e3b63df4c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1bf87aa657e278e3bca4d1721c2c7d87d83ce64748d40346a2ec287bf0d6b118", "generation_seconds_batch": 20.67536067601759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09168246388435364, "target_z": -0.42382811430573886, "ce": 5.990842819213867, "genuineness": 4.645420074462891, "identity": 0.809374213218689, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1f3746761533cd6d7264cd9a3db7d4aca8a132aa1b0c1a206bafdd3e3b63df4c.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "1f629631414060954e88f27457e7ea24238561a06e66a1c39f28871df0b67a2e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "53a7e7f64114dea479a16866298689723a6888aff7b731eb91cf0ad8dc86e77a", "generation_seconds_batch": 28.631439635995775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38156405091285706, "target_z": 0.11962889331867399, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.2688308358192444, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1f629631414060954e88f27457e7ea24238561a06e66a1c39f28871df0b67a2e.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "1f7b796a7aa5505612fe5979d977403f2d160e3ce02fc0b9b13111f98897d417", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bff8a8a804f3062d2b9ab1ad2a027667fbe5d9c0be54ae48a35f21e4ed59f6ba", "generation_seconds_batch": 16.02757140796166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.384690046310425, "target_z": 3.874999734607723, "ce": 5.649308204650879, "genuineness": 4.469690799713135, "identity": 0.633399248123169, "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/emo_Confusion/1f7b796a7aa5505612fe5979d977403f2d160e3ce02fc0b9b13111f98897d417.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "1f86f3749dc8b9992932362251dcf394af6427b239eb9e4e17650c78f6e77b78", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e2e547f1aa28cb55d1d69ff439f528f3c2159ca904e658066a0bd63931dddee9", "generation_seconds_batch": 16.088130385032855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05938658118247986, "target_z": -0.4843749877779872, "ce": 5.982900142669678, "genuineness": 4.68307638168335, "identity": 0.4388241469860077, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/1f86f3749dc8b9992932362251dcf394af6427b239eb9e4e17650c78f6e77b78.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "202b6ff4628b499b540ce7cc99c76f8c047f3f9c8f5e451790bda2d4825efdc2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "90f03ab0fb7305822285f46691c1942810d202d2ff2ac5c1723c21ee15eb7ca8", "generation_seconds_batch": 15.590884095989168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8303205966949463, "target_z": 0.9609375532530556, "ce": 5.510311126708984, "genuineness": 4.494794845581055, "identity": 0.3321811556816101, "hypothesis": "Die Tür ist noch offen. Ich hab dein 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/emo_Confusion/202b6ff4628b499b540ce7cc99c76f8c047f3f9c8f5e451790bda2d4825efdc2.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "203c8a6f2025151fde0d36c626e1e5ea3bc09fc3e261cd63718fd560ae9188bd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47b33a57b7726cc9871eeea03b0287ca60b4b2bc6e70495511d2e99dffa03274", "generation_seconds_batch": 16.289375386899337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1678646802902222, "target_z": 1.5937500733320766, "ce": 5.266074180603027, "genuineness": 4.137060642242432, "identity": 0.7225006818771362, "hypothesis": "J'ai trouvé la petite boîte à buite. 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/emo_Confusion/203c8a6f2025151fde0d36c626e1e5ea3bc09fc3e261cd63718fd560ae9188bd.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "2110517004c7aded791012ddecbf302d15438e40997ba537537ad5d4286b93e3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9a1df25a478365550ba5db71ff828504326d653d342b8967cf9e4033111510ec", "generation_seconds_batch": 20.108450368978083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.226205587387085, "target_z": 1.7031249912699908, "ce": 5.700935363769531, "genuineness": 4.607763767242432, "identity": 0.7111762166023254, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2110517004c7aded791012ddecbf302d15438e40997ba537537ad5d4286b93e3.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "21bba435e9b532bd48447244cbbbed5e0423626a79a8d5ee3d24b53c40a84294", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "144af2432961185105a410b1e026271d7dd28abfc7060651a4cda130a5480238", "generation_seconds_batch": 20.968235336011276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12397834658622742, "target_z": -0.36328124083349045, "ce": 5.911416053771973, "genuineness": 4.8588056564331055, "identity": 0.7097307443618774, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/21bba435e9b532bd48447244cbbbed5e0423626a79a8d5ee3d24b53c40a84294.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "21d0be487408908be849c3ec2407aac527c91c159ac5cdb785b4440cbacb57ee", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5233b8eaefbc1631478bf3d051a543347eb203307a6acdc8efc525413245e54e", "generation_seconds_batch": 25.231844969908707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11981113255023956, "target_z": -0.3710937469444968, "ce": 5.879645347595215, "genuineness": 4.770940780639648, "identity": 0.8090646266937256, "hypothesis": "J'ai trouvé la petite boîte Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/21d0be487408908be849c3ec2407aac527c91c159ac5cdb785b4440cbacb57ee.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "22131484091db3d59c64e931d62a3093d9643b4c121a654dc6b126adb51140fa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26fd94c7970085b028de5f31082d13a26f22bb37eeeb0a13a8ae01a20b687374", "generation_seconds_batch": 18.85379116097465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4479789733886719, "target_z": 0.2441406134327379, "ce": 5.919358730316162, "genuineness": 4.20609712600708, "identity": 0.33790159225463867, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/22131484091db3d59c64e931d62a3093d9643b4c121a654dc6b126adb51140fa.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "22412dcd4caefbf756b3c21eb35caeaa5814d8d607decb44a979342fa31a49de", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "28fc67d052740e69fd4fb1be37af9896f37f217cb5639d0c7cf454c323825c01", "generation_seconds_batch": 22.32911538798362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00104561448097229, "target_z": -0.5937500174600182, "ce": 6.300607204437256, "genuineness": 4.218649387359619, "identity": 0.4967195987701416, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinschauen.", "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/emo_Confusion/22412dcd4caefbf756b3c21eb35caeaa5814d8d607decb44a979342fa31a49de.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "2276e5a925c70ae60676fae5738da9bff5bfaa36e017416bf42cdf389d1b7c81", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5de068166b7e31f09dd1a13a2cf677165823c59ddc5a2c4b43b41bc7178d5e31", "generation_seconds_batch": 22.078123525949195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5013720989227295, "target_z": 4.093750017460018, "ce": 5.764476776123047, "genuineness": 4.055472373962402, "identity": 0.6226340532302856, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2276e5a925c70ae60676fae5738da9bff5bfaa36e017416bf42cdf389d1b7c81.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "2298846044c3f38aa98c405dcb89889abeb4128af2964eb6745b9b93a05c3e4d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1d93b69dcdf62eae12fb74162e3efcfa8f16cbb02a911a733367afda4631a304", "generation_seconds_batch": 25.46120472298935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0595171451568604, "target_z": 1.390624970317969, "ce": 5.498397350311279, "genuineness": 4.432034492492676, "identity": 0.5725303888320923, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2298846044c3f38aa98c405dcb89889abeb4128af2964eb6745b9b93a05c3e4d.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": "230fdcc73eefb1991d64fc9f06b65c71fd77ee4e134ca209740550b5db421697", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6c22b5ccdecb6fcd23aff64905dc26745d55c58be72a9d31db694d67d154865", "generation_seconds_batch": 22.034487597993575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0928548574447632, "target_z": 1.45312501920602, "ce": 5.546053409576416, "genuineness": 4.720732688903809, "identity": 0.6845399141311646, "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/emo_Confusion/230fdcc73eefb1991d64fc9f06b65c71fd77ee4e134ca209740550b5db421697.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "232737e312f8db22fc2427a2fcd8e03c909b10677f969175919b9f9793aaeb1f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e931753554c9908cd36a96ec1091ec4f158018971279b9c83b161a32579ec0ea", "generation_seconds_batch": 20.611866575898603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9094976186752319, "target_z": 1.1093750855540894, "ce": 5.4388275146484375, "genuineness": 4.256305694580078, "identity": 0.5389778017997742, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/232737e312f8db22fc2427a2fcd8e03c909b10677f969175919b9f9793aaeb1f.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": "2427bfbaebb79daa2439cdd1c3422536192628d5eac7a657dc07b724d90881a7", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f831f8b834b6279e1d0c3b74357e2ad7562afdbc9aefad6d8bebde9831c3cdbe", "generation_seconds_batch": 18.491003404953517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7096021175384521, "target_z": 2.6093751414261477, "ce": 5.808161735534668, "genuineness": 4.469690799713135, "identity": 0.6128125786781311, "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/emo_Confusion/2427bfbaebb79daa2439cdd1c3422536192628d5eac7a657dc07b724d90881a7.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "25a77dbb3536bc23321f495d240c2644c6f8493d68bdb9882628169dc8bb7295", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6e51b04b4092aa375b43d93e1614e343f2f8a8eb4d3ee1eec3e33ac9cf1b1d3b", "generation_seconds_batch": 14.635950949974358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13231275975704193, "target_z": -0.3476562565475068, "ce": 5.879645347595215, "genuineness": 4.256305694580078, "identity": 0.645980179309845, "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/emo_Confusion/25a77dbb3536bc23321f495d240c2644c6f8493d68bdb9882628169dc8bb7295.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "25b03ae4a673f908dd0486140e739733e05e33cf9284038f722c5e4a607d3162", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6f85bb576b2da7a5c0218ebd07149d748fbe8936d0fcea864312f887483c0524", "generation_seconds_batch": 19.924826035043225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0303466320037842, "target_z": 1.335937399604895, "ce": 5.704906940460205, "genuineness": 4.545003414154053, "identity": 0.7705236077308655, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/25b03ae4a673f908dd0486140e739733e05e33cf9284038f722c5e4a607d3162.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "25d218e86c235e88e61ed745a2baad2b9361fb7e9bb54bd47e50b199d8dbae0c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a5586a1d6cf5a836eacef919205354b45783350d9d157149547a027dfca71825", "generation_seconds_batch": 20.618317262036726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9679691791534424, "target_z": 3.093750129204135, "ce": 5.839932441711426, "genuineness": 4.18099308013916, "identity": 0.48229968547821045, "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/emo_Confusion/25d218e86c235e88e61ed745a2baad2b9361fb7e9bb54bd47e50b199d8dbae0c.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "26049253a27d3c9dbd0cca61e6e85dc6d76ea258dfd0cefce1bbeaaa0c9c779b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7ee8af71d058b56f33abd1a24afc14ebe2facd6c654cebab96a3d53a436fdaf2", "generation_seconds_batch": 21.051624797051772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6386288404464722, "target_z": 0.6015624397629371, "ce": 5.824047088623047, "genuineness": 4.168440818786621, "identity": 0.5708296298980713, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/26049253a27d3c9dbd0cca61e6e85dc6d76ea258dfd0cefce1bbeaaa0c9c779b.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "2684e3f4f3db88eced8fda59cdac1fd9843229b6fbbbdc0c5231ba2fe7fa221e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "539b443bbbdfa1467376317f63367565ba9403700a5e34df252451693086ca87", "generation_seconds_batch": 21.233912243973464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09689147770404816, "target_z": -0.41406248865098816, "ce": 6.356205940246582, "genuineness": 4.570107460021973, "identity": 0.34402114152908325, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2684e3f4f3db88eced8fda59cdac1fd9843229b6fbbbdc0c5231ba2fe7fa221e.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "26a1baf53e1790e1433342a7755cf4abb8afdc9f61d655f6f71d0be605865657", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "98c9a16228157ac8ca661deadf6355c1c350535cf7778592e07292e5423d6946", "generation_seconds_batch": 21.928549460950308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08959884941577911, "target_z": -0.4277343813292566, "ce": 5.990842819213867, "genuineness": 4.494794845581055, "identity": 0.5188325643539429, "hypothesis": "Ik 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/emo_Confusion/26a1baf53e1790e1433342a7755cf4abb8afdc9f61d655f6f71d0be605865657.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19c3f28739385380646c09c1d8d1c76d24257b2b9aca177fcf4aeb402c6594ce", "generation_seconds_batch": 19.701358182006516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.395919322967529, "genuineness": 4.162164688110352, "identity": 0.3284386992454529, "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/emo_Confusion/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "2724fc57b3885555039deb8af85a7c419dd1258d2728478136ed8a320ac36bae", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e9ca69fba746ed818ab6e289b3876232923c144e27f26f7ad0cbd8dc02bfc852", "generation_seconds_batch": 21.086008926969953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2178711891174316, "target_z": 1.6875000349200364, "ce": 5.609594821929932, "genuineness": 4.720732688903809, "identity": 0.6998053789138794, "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/emo_Confusion/2724fc57b3885555039deb8af85a7c419dd1258d2728478136ed8a320ac36bae.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": "27329ace7b83bae0c9c1cdeff19b78a8d0dbe94fe4b71df83a089d9180dada6c", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "cd972c60e4f83af2bb1c6bd3268275b5a9ab9087f57a5b33869eb19890fa49bd", "generation_seconds_batch": 20.5698393820785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.976172924041748, "target_z": 1.234374959841958, "ce": 5.796247482299805, "genuineness": 4.720732688903809, "identity": 0.788463830947876, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/27329ace7b83bae0c9c1cdeff19b78a8d0dbe94fe4b71df83a089d9180dada6c.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "276b7ac601bd59521f9e87ef9fd8431590ab1f744d60aeb200829323d803c233", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "151f2c3c6959a169816918bcf5fbf57be8bc474134a904017a05d9cf4840f1bf", "generation_seconds_batch": 21.815699794096872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.401228427886963, "target_z": 2.0312499685719674, "ce": 5.34748649597168, "genuineness": 4.432034492492676, "identity": 0.6002480387687683, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/276b7ac601bd59521f9e87ef9fd8431590ab1f744d60aeb200829323d803c233.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": "278ca49e33ab4deb358c1d7ff43ffa92538523c63404d553f797366d39aca9fb", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d95e4eecc3b3743758ba0cf46f9400a600d609ff1d4818968452ca6128bac363", "generation_seconds_batch": 24.104134701075964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.482358455657959, "target_z": 0.3085937259924749, "ce": 6.0305562019348145, "genuineness": 4.344170093536377, "identity": 0.5323152542114258, "hypothesis": "I wrote the number on the envelope. Rate it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/278ca49e33ab4deb358c1d7ff43ffa92538523c63404d553f797366d39aca9fb.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "27bbb6e2eac87d809f86841b00e50cfdba12814447da769d211e72e387f599ff", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "60d3c05ea321cd4d9368d134b0a883cc84d32fcdb4d92a6e3114ac891ccafe30", "generation_seconds_batch": 20.714532652986236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.030208468437194824, "target_z": -0.6523437713885223, "ce": 5.808161735534668, "genuineness": 4.369274139404297, "identity": 0.5632796883583069, "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/emo_Confusion/27bbb6e2eac87d809f86841b00e50cfdba12814447da769d211e72e387f599ff.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "27ef13f77594cfa75e81a082b239a93f1465339566613db602c72fc6cda80f5a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ae954dbacc31d0959d8def77ceff9ba5f789c391ae7e67de083259da3ccc971f", "generation_seconds_batch": 22.191772917984053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1823192983865738, "target_z": -0.2539062390874886, "ce": 6.395919322967529, "genuineness": 4.607763767242432, "identity": 0.25603169202804565, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/27ef13f77594cfa75e81a082b239a93f1465339566613db602c72fc6cda80f5a.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "27ff07009dfd916b5af8e22e8b2b598145a34ca444ce7346d02c837314311fd3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "88f611ac35d780df43ea25b67152d44ae00e95cbe275945e52554991d41c7eb6", "generation_seconds_batch": 21.106531191035174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40213966369628906, "target_z": 0.15820313001975525, "ce": 6.0146708488464355, "genuineness": 4.105680465698242, "identity": 0.6624072194099426, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/27ff07009dfd916b5af8e22e8b2b598145a34ca444ce7346d02c837314311fd3.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": "28266a347d539424a6ad5456d33b4396dacdba1972485d615eb57a126badd6db", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1800cd2c64374c769a125113f8544513636d0fa32a93ca1bdfb814d5f018128e", "generation_seconds_batch": 21.846888614003547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1428613662719727, "target_z": 1.54687498079398, "ce": 6.197352409362793, "genuineness": 4.293961524963379, "identity": 0.427823543548584, "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/emo_Confusion/28266a347d539424a6ad5456d33b4396dacdba1972485d615eb57a126badd6db.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": "284046a03297a36b9f115c679b7b5735e109959a483bd425b45a6247cebd0eed", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "266f6cf233ff1087dc079258b5686dc2df83e91d1c503201f8c3d32560086c46", "generation_seconds_batch": 21.95666484697722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5907059907913208, "target_z": 0.5117188290065825, "ce": 5.998785495758057, "genuineness": 4.281409740447998, "identity": 0.41465237736701965, "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/emo_Confusion/284046a03297a36b9f115c679b7b5735e109959a483bd425b45a6247cebd0eed.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "2888a85c595f3af0cd78dbdc13b4c8e28c261d8dd6544640105d37e95bb58993", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e550a7279f151817fe39193233121dcf9b59efad87961e960159c040dee2cb9b", "generation_seconds_batch": 16.004264255054295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08543163537979126, "target_z": -0.435546887440263, "ce": 5.796247482299805, "genuineness": 4.632867813110352, "identity": 0.7152935266494751, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2888a85c595f3af0cd78dbdc13b4c8e28c261d8dd6544640105d37e95bb58993.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": "289140e1f08d4751b7843c93ce9490ed3849bf05f7c90822a5d102e4e4e6633a", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bafc5f72a2f0155253275af9e99f4871a7851ad54c53213ce834fc8e8bbebd92", "generation_seconds_batch": 18.94644432491623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0929853916168213, "target_z": 3.328124921429918, "ce": 5.569881439208984, "genuineness": 4.118232727050781, "identity": 0.39918142557144165, "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/emo_Confusion/289140e1f08d4751b7843c93ce9490ed3849bf05f7c90822a5d102e4e4e6633a.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "28ecaeb54e9868ac2779961f64b6143fc0ebef115f395b2c37f0f853ba9c4c81", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c4f39cd1d595943458d2a544a8a3079deb3f35a476f6f4f0f1f35eefe4ab37be", "generation_seconds_batch": 19.373244078014977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7012676000595093, "target_z": 2.59374996158796, "ce": 5.645336627960205, "genuineness": 4.394378662109375, "identity": 0.7830898761749268, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/28ecaeb54e9868ac2779961f64b6143fc0ebef115f395b2c37f0f853ba9c4c81.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": "29187d805950e2ecb07d0ca9ee1cad088d15d57d02d3270775e263cee0a89d3b", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c19b6762475d87f19f784f9e686c4e5d9b8b6f27f32db35b275e79068168c953", "generation_seconds_batch": 21.928311440977268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6429266929626465, "target_z": 2.484375043650046, "ce": 5.967014789581299, "genuineness": 4.65797233581543, "identity": 0.8571401834487915, "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/emo_Confusion/29187d805950e2ecb07d0ca9ee1cad088d15d57d02d3270775e263cee0a89d3b.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "2950a4aa3eefc872f44872afbae421e055d2ff07f4d26a4e76ebd20da4b4caf6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "925bec8e7d56d3f4f69de4f814e194d7e312297595ea8cdcd940cee6b6b800da", "generation_seconds_batch": 21.928588861017488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1271037459373474, "target_z": -0.3574218822022575, "ce": 6.09409761428833, "genuineness": 4.20609712600708, "identity": 0.5730893611907959, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2950a4aa3eefc872f44872afbae421e055d2ff07f4d26a4e76ebd20da4b4caf6.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "2a32e1586112f6eb8684bacc2e899716b6994ae51e091eb15c573606a5aa69fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2f1ab44e75764dfdab138ed4370897e65dc477705bd0f6b684d009eabd0758dd", "generation_seconds_batch": 22.04128434404265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 5.831989288330078, "genuineness": 4.469690799713135, "identity": 0.5518070459365845, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2a32e1586112f6eb8684bacc2e899716b6994ae51e091eb15c573606a5aa69fe.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "2ad5a775d641c436d10d848de559e0f9d9e5b1deb81a2dbb11f42d3c0135c14b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a2a43bb46c4f90e72d70a46d389b0a844ffa5c7a272c1f91be2ca10dbadb565b", "generation_seconds_batch": 20.96034846501425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.368021249771118, "target_z": 3.843749821907814, "ce": 5.013894557952881, "genuineness": 4.218649387359619, "identity": 0.7170896530151367, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2ad5a775d641c436d10d848de559e0f9d9e5b1deb81a2dbb11f42d3c0135c14b.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "2be6759daaead4163ef95d6cf36ad55c910d3dcf69e72c47afc5c090d066ff11", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5832c00572c621c0d1a1f73a86966e9b3da31cc90e5756b855e8bb2199f879de", "generation_seconds_batch": 24.316069032996893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05313575267791748, "target_z": -0.4960937609125114, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.3859647810459137, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2be6759daaead4163ef95d6cf36ad55c910d3dcf69e72c47afc5c090d066ff11.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": "2c26ee43cec4c9cf8861350ebfc96e2d5b809c60b41023f46b23fc69fca10d4b", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7dba922c270c81555731a6d3fb979cf41379ba47ca06943c6ab34196bbcb826", "generation_seconds_batch": 29.14194281003438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417897343635559, "target_z": 2.0625001047601095, "ce": 5.458683967590332, "genuineness": 4.375550270080566, "identity": 0.5406007766723633, "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/emo_Confusion/2c26ee43cec4c9cf8861350ebfc96e2d5b809c60b41023f46b23fc69fca10d4b.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "2cdc681f0531b1f33be42aab0a97e57a38e3750479f00aa9473e57d5e0ef379e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4c327aea9e685b20530a587251b45c689b8eaf5c384980ab68a9d710bc5681c4", "generation_seconds_batch": 21.902440170990303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9094976186752319, "target_z": 1.1093750855540894, "ce": 5.998785495758057, "genuineness": 4.720732688903809, "identity": 0.58636873960495, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2cdc681f0531b1f33be42aab0a97e57a38e3750479f00aa9473e57d5e0ef379e.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "2d0cc220ea6942b0b0c07e007079ed1eb40bb4de44f43d00599b57b36b4e9327", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a101efd5e8d866e886cecf5f2023ac13b8c7b6511cc983ea02948aed17714793", "generation_seconds_batch": 20.042186809005216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8969959020614624, "target_z": 1.0859374275409244, "ce": 5.919358730316162, "genuineness": 4.2751336097717285, "identity": 0.47785332798957825, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2d0cc220ea6942b0b0c07e007079ed1eb40bb4de44f43d00599b57b36b4e9327.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "2d7c0e2e002ed914ba43620567ff51ae2af3d317ecef78b3a969c0828fb0e17c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b4c440aa29884194abd073031759ccc8a744ba29c89a5c264734f79cdb80556", "generation_seconds_batch": 14.786553442943841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8678255081176758, "target_z": 1.0312500803160838, "ce": 5.585766792297363, "genuineness": 4.7332844734191895, "identity": 0.536091685295105, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2d7c0e2e002ed914ba43620567ff51ae2af3d317ecef78b3a969c0828fb0e17c.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": "2dbd7c9708914353a2ae1790b868887dc51b019d0b189dfe82297078af210314", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "aa1cf59fb5719f65c298334787e921e9a5f1e650cd4d9d2e65869ec34d3d1c19", "generation_seconds_batch": 15.828915882972069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36932286620140076, "target_z": 0.0966796688396055, "ce": 5.788304805755615, "genuineness": 4.256305694580078, "identity": 0.5829554796218872, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I it will meet you on the other side.", "wer": 0.041666666666666664, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2dbd7c9708914353a2ae1790b868887dc51b019d0b189dfe82297078af210314.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "2eed48331517f9a8ccf6ee7e01d7c84d1afe5b8d639c526b8fc351e4ba55f10b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9466b79cc532feb3c803266bee1cbd7c6063bd0b4343124b08828a46e41bc3f4", "generation_seconds_batch": 14.691540786065161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.6912357211112976, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2eed48331517f9a8ccf6ee7e01d7c84d1afe5b8d639c526b8fc351e4ba55f10b.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "2f2ef8a638777419ceed2fe5d524632b5e9e1ecc11711b90c935e784548dbfa2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8e7edcc256fa898e6300add83db46a5f334d4a17e56f1aa2b3f421faf9e3c45d", "generation_seconds_batch": 21.398158353054896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.301215410232544, "target_z": 1.8437500453960474, "ce": 5.927301406860352, "genuineness": 4.18099308013916, "identity": 0.6577948331832886, "hypothesis": "De deur 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/emo_Confusion/2f2ef8a638777419ceed2fe5d524632b5e9e1ecc11711b90c935e784548dbfa2.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "2f33de9a9494dcd78507b8fb5596b0687962ee302e672e1a4146cfc9fefe9497", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0f0afb2bcf301274ec7f1d7e1f2d2b540ee58e1055f01236c2a386263ae4c521", "generation_seconds_batch": 21.051624797051772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7990664839744568, "target_z": 0.9023437434524931, "ce": 5.967014789581299, "genuineness": 4.331617832183838, "identity": 0.6396398544311523, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2f33de9a9494dcd78507b8fb5596b0687962ee302e672e1a4146cfc9fefe9497.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": "2fab36211b9ba6eed2d617b9e9fc4d388f8a9744deba5eada24a287826357f11", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab460fbed5bdc49327482a9328417156533499cb8150ee7d07e098ba5d8bbac1", "generation_seconds_batch": 14.019105428946204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7347357273101807, "target_z": 4.531249689211675, "ce": 5.4666266441345215, "genuineness": 4.068024158477783, "identity": 0.42245450615882874, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2fab36211b9ba6eed2d617b9e9fc4d388f8a9744deba5eada24a287826357f11.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "2fab418e943a412f90aaea7af1ee96063eb15cd398583942dedd02d92c659f77", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "137efb0f60badd7fa8bbfe79c69121ea24d3882a833682d655fadadc5e98d6d4", "generation_seconds_batch": 24.631402708939277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 5.879645347595215, "genuineness": 4.632867813110352, "identity": 0.7527270317077637, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2fab418e943a412f90aaea7af1ee96063eb15cd398583942dedd02d92c659f77.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": "2fb7ad8275d5e8ebbf873244d8aa3f158050ec9a9c61cb60448d1e8b9ef675e1", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1651fb61d2a323c5c2b15db9baf81ae72687f99e42c9c83eb0550c4e27130cb6", "generation_seconds_batch": 21.457923726993613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.318014621734619, "target_z": 3.7499996368316206, "ce": 5.871703147888184, "genuineness": 4.406930446624756, "identity": 0.6154951453208923, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/2fb7ad8275d5e8ebbf873244d8aa3f158050ec9a9c61cb60448d1e8b9ef675e1.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "30a6cf23daf2eefd840186ef0b90199ab8f334147b59ecda4d7cb03bbfb6df9d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e11260cb3bd3f4c6863376937fc792614840508ce8a0c5c049547b592e378c4c", "generation_seconds_batch": 22.400838118046522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16981765627861023, "target_z": -0.27734375742050776, "ce": 5.808161735534668, "genuineness": 4.545003414154053, "identity": 0.7193671464920044, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/30a6cf23daf2eefd840186ef0b90199ab8f334147b59ecda4d7cb03bbfb6df9d.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": "31288329b95be742cf31c9962cc4ea375e2fcacf7e8bab951dcaa328be17766e", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "190e4d902147bae2d8ac81df253582f921b4add9fbe5444bd66ce3fddf140fc1", "generation_seconds_batch": 16.361110532074235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2845464944839478, "target_z": 1.8124999092079053, "ce": 5.974957466125488, "genuineness": 4.344170093536377, "identity": 0.32542917132377625, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/31288329b95be742cf31c9962cc4ea375e2fcacf7e8bab951dcaa328be17766e.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "31bfd2cd18f47301f74a71473cd203861ca7d97c58b6e6a71fb39986036646e0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d5916cabb292388b16cf2d06aee17036e98ae3b5b09a88d43ce182e5f6ae9f80", "generation_seconds_batch": 19.679027174017392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0220122337341309, "target_z": 1.3203124432549407, "ce": 5.538110733032227, "genuineness": 4.507347106933594, "identity": 0.7171055674552917, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/31bfd2cd18f47301f74a71473cd203861ca7d97c58b6e6a71fb39986036646e0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "325253f834f7ef27446a40723bb4bc55e4b0300a875023986ca8dbaa1304c502", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "db326397685ff8db2f7d0d38b23d85adb8c64ec2bff3176075e605534a2fa9d3", "generation_seconds_batch": 21.80175695195794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6990534067153931, "target_z": 0.7148437085324567, "ce": 5.831989288330078, "genuineness": 4.65797233581543, "identity": 0.6785698533058167, "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/emo_Confusion/325253f834f7ef27446a40723bb4bc55e4b0300a875023986ca8dbaa1304c502.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "325bcb0108f15cc375761a5448e9f7e534168a06702e37c5e694a4352bbcbd44", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b7ac423f5299444c506560e9a21e5a140f323b36abb094908279431d36d4708d", "generation_seconds_batch": 21.553457865025848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.384690046310425, "target_z": 3.874999734607723, "ce": 5.450741291046143, "genuineness": 3.8044304847717285, "identity": 0.33430004119873047, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/325bcb0108f15cc375761a5448e9f7e534168a06702e37c5e694a4352bbcbd44.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "3282e66b3c9bd433a4ed499aa7119957647d9df016cb86e1d2e914c7f443615b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70ccb661a10b8700707e88c1f7ec9711f5355cf97cefc142b1ba581c12f295dd", "generation_seconds_batch": 21.634572334005497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6573812961578369, "target_z": 0.6367187032944512, "ce": 5.522225379943848, "genuineness": 4.080576419830322, "identity": 0.30823615193367004, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3282e66b3c9bd433a4ed499aa7119957647d9df016cb86e1d2e914c7f443615b.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "329f53906b1693d28d28217772b1ca30adba526e7533517d54bff8e08af2e0b1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b4eccaa42207cde87fc87d9b166661506028f9f0830e759006791f1356876e7f", "generation_seconds_batch": 15.628392399055883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22034507989883423, "target_z": -0.18261720812464655, "ce": 5.561938762664795, "genuineness": 4.344170093536377, "identity": 0.8230074644088745, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/329f53906b1693d28d28217772b1ca30adba526e7533517d54bff8e08af2e0b1.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "32e86a4db1b8261cf1b7009d6c4cd31c73c1d78d4c617101ae7db0c09eb4d920", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c4a1b366cd66e57dd37ae10f8cc9f8691d2331c9a81dc66484e156d0ee4f7da4", "generation_seconds_batch": 20.305700591066852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04480132460594177, "target_z": -0.5117187731345242, "ce": 6.0305562019348145, "genuineness": 4.2751336097717285, "identity": 0.7195436954498291, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/32e86a4db1b8261cf1b7009d6c4cd31c73c1d78d4c617101ae7db0c09eb4d920.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": "32f82b7198365cf42f26686c2f195cae68fc5a738857e99b0c7485b0f6b27780", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09da70e8f858561efe78b5e80a7b0e3a2e19e882130b435438022f25c1659158", "generation_seconds_batch": 15.9259085690137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23024222254753113, "target_z": -0.16406248865098816, "ce": 5.967014789581299, "genuineness": 4.337893962860107, "identity": 0.30645304918289185, "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/emo_Confusion/32f82b7198365cf42f26686c2f195cae68fc5a738857e99b0c7485b0f6b27780.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "3341f88f137f601050bbf9c1f1cac8dedb1297780c439270053970b9c49c0101", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ba76cdc8e6bfc0a766c1701809fabef75bf1471c229703e899ce6b9f787e9f73", "generation_seconds_batch": 21.208438965957612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 5.959072113037109, "genuineness": 4.796044826507568, "identity": 0.590961754322052, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3341f88f137f601050bbf9c1f1cac8dedb1297780c439270053970b9c49c0101.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": "335d4839605a751c8eb651405fa4b5a7f446cefd0fd4403d2e7bbc1ba91958b4", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7c19aa014fc69fc8f20bfb3734962cb014b0a850697c86ca1d15221e8bdebd0d", "generation_seconds_batch": 21.139035118976608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8929593563079834, "target_z": 2.9531250750780784, "ce": 6.0305562019348145, "genuineness": 4.381826400756836, "identity": 0.11847292631864548, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/335d4839605a751c8eb651405fa4b5a7f446cefd0fd4403d2e7bbc1ba91958b4.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": "337e6f5e1f6626569b0fc5f378996a1c9dc42421739863eff7ee0babcb342dcf", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a727c330e2fa45afc08dca165b72d7194c4ff9bb24ef7cc1a2a555797cdc4330", "generation_seconds_batch": 21.623937588068657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417897343635559, "target_z": 2.0625001047601095, "ce": 5.323658466339111, "genuineness": 4.080576419830322, "identity": 0.5062988996505737, "hypothesis": "La route tourne à gauche près 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/emo_Confusion/337e6f5e1f6626569b0fc5f378996a1c9dc42421739863eff7ee0babcb342dcf.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "33cff3620771b92960b294fcdca24718d078efcbbfd505e5532d605bfada537c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d35ca8f32cbed9aa29037e9b2f29a74fdb0039125ced9ed252b5b1f53d1c8133", "generation_seconds_batch": 26.24704099202063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14064718782901764, "target_z": -0.3320312443254941, "ce": 5.498397350311279, "genuineness": 4.632867813110352, "identity": 0.6570927500724792, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/33cff3620771b92960b294fcdca24718d078efcbbfd505e5532d605bfada537c.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "4ab1a49d65ed6c7ea30f12be9c36a2a473bcec99d016b06546b2e09da612dab1", "generation_seconds_batch": 16.312376318965107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3928940296173096, "target_z": 2.0156250122220127, "ce": 6.070269584655762, "genuineness": 4.375550270080566, "identity": 0.6318168640136719, "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/emo_Confusion/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "33f866dd51086c89d657d5e69702b6bb4e61dda45cc7d021d7d99dbefe6be6b1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e6330de7734e537808567b42e97bc95eb4346b4d562ff0458a0b4c90ef058c14", "generation_seconds_batch": 15.944600107031874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6698829531669617, "target_z": 0.6601562495634995, "ce": 5.855817794799805, "genuineness": 4.582659721374512, "identity": 0.6671287417411804, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/33f866dd51086c89d657d5e69702b6bb4e61dda45cc7d021d7d99dbefe6be6b1.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": "3443e7f4056f3814f25e15cc4a7bb333f5daa0b791d3fa91b9325ace86052a52", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89079fb075a7eae8a80ed6ece378a42c0e5e1c9330925e71e248f2772429d6bc", "generation_seconds_batch": 22.580438681994565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8929593563079834, "target_z": 2.9531250750780784, "ce": 5.585766792297363, "genuineness": 4.400654315948486, "identity": 0.03656135872006416, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine Houtenek. Ik wacht aan de andere kant.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3443e7f4056f3814f25e15cc4a7bb333f5daa0b791d3fa91b9325ace86052a52.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "16faa148398df34087197461246b1015a466e85bbea9e2171eee4cbeacc9dcc7", "generation_seconds_batch": 16.154919733991846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017714470624923706, "target_z": -0.5624999930159927, "ce": 6.348263263702393, "genuineness": 3.8295347690582275, "identity": 0.6041673421859741, "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/emo_Confusion/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.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": "348726771fec2e9c7d7c5770e0a5c96483d251620c196ffc87ab414a55a49773", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ae53d874ebcac78af0f8df8d2f89aeb58bf63c681dcc28066007f2f4c0ac457", "generation_seconds_batch": 20.067378230043687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417897343635559, "target_z": 2.0625001047601095, "ce": 5.839932441711426, "genuineness": 4.13078498840332, "identity": 0.5302870869636536, "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/emo_Confusion/348726771fec2e9c7d7c5770e0a5c96483d251620c196ffc87ab414a55a49773.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "34bdb33ea21c26364b8886d1b685ee1c3c7537e69a56656322cce8b21cf98897", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cd3a548ab7254ee0a32f45899cd88f93d3913f477d6bd9bd6b9b8224f9a24007", "generation_seconds_batch": 20.502132874098606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.692367434501648, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/34bdb33ea21c26364b8886d1b685ee1c3c7537e69a56656322cce8b21cf98897.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "34d79c5cbcdc90c3982ab97890aaf4e7d7943b92a134568b98e9c6e0c00fdd78", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b4b3a166b445adfe0f6d571dd172891ae5e19647c1c1c404abe5eecfeab997b8", "generation_seconds_batch": 21.773904109024443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.482358455657959, "target_z": 0.3085937259924749, "ce": 6.0305562019348145, "genuineness": 4.6956281661987305, "identity": 0.6553327441215515, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/34d79c5cbcdc90c3982ab97890aaf4e7d7943b92a134568b98e9c6e0c00fdd78.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "35461e03ef6f84d42fe874cba1d67ab9537a73c35a4560c10950a797380c5c0b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1adfbce63aea55950f0f1ebac46b2c8d8e33a4b6364d8702c0e706574c922dbc", "generation_seconds_batch": 19.06448508996982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.517910361289978, "target_z": 2.250000027936029, "ce": 5.839932441711426, "genuineness": 4.262581825256348, "identity": 0.5037811398506165, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/35461e03ef6f84d42fe874cba1d67ab9537a73c35a4560c10950a797380c5c0b.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "3566f5964f63f283436992f3f040f84f722b7b677ff37946b948e7450897a900", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b8bad7a279de25d4c93fd9df7a30b7343baa44511e8e9940bb4a915081a51179", "generation_seconds_batch": 21.53755801601801, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5115289688110352, "target_z": 0.36328129670554876, "ce": 5.526196479797363, "genuineness": 4.545003414154053, "identity": 0.6885765194892883, "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/emo_Confusion/3566f5964f63f283436992f3f040f84f722b7b677ff37946b948e7450897a900.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "356a4060318a9d72f7457a439593de0a78c530559c4fc197a9e1daea7967db9f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e90159c6ec2245145d5ebf7a8dfd9b79f5479d3a8de87272dab8b8ccf2429d0e", "generation_seconds_batch": 20.593501407071017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47194045782089233, "target_z": 0.2890625305550319, "ce": 6.109982967376709, "genuineness": 4.545003414154053, "identity": 0.538104236125946, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/356a4060318a9d72f7457a439593de0a78c530559c4fc197a9e1daea7967db9f.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "35ba5ac52ead6526edc7da43286e94008dbf38ed375174ae2a24a4eb4c876949", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a66a3e87ff3afc7085903cf274887f5fa85e85d14525a2e142f39278eee0b34", "generation_seconds_batch": 20.73501576099079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7094714641571045, "target_z": 0.7343750157140164, "ce": 5.669164657592773, "genuineness": 4.607763767242432, "identity": 0.6661235094070435, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/35ba5ac52ead6526edc7da43286e94008dbf38ed375174ae2a24a4eb4c876949.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "36a5fe2348696e81d4c013f79958779780c8698c2f56662c08106aef52df6ae9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "eea3f4eb1f67da60ff624605f79d22b790522c68a4ad16bcf993c0bf73b4ed53", "generation_seconds_batch": 15.659956371993758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6345922946929932, "target_z": 2.468750087300091, "ce": 5.601652145385742, "genuineness": 4.469690799713135, "identity": 0.6028555035591125, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/36a5fe2348696e81d4c013f79958779780c8698c2f56662c08106aef52df6ae9.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "36f82a2a0c97cb91a1ba081dfa162af7093121bf3903cc6e5c1c817b39b77eb5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d573dfcf791409df66d2f1ecc7fea1a91a5b5455ebc0c62fd075c7294b5f7e64", "generation_seconds_batch": 15.820022452040575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8762905597686768, "target_z": 2.9218751623781696, "ce": 5.518253803253174, "genuineness": 4.68307638168335, "identity": 0.5423456430435181, "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/emo_Confusion/36f82a2a0c97cb91a1ba081dfa162af7093121bf3903cc6e5c1c817b39b77eb5.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "37595508bf581342e3a2279047da35ce78c7819b198c55438755295bffe542a5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e86ce6d5ae79bbfcf3ac699909ebe6716195319c0cbe9b28a25304acd3a6333e", "generation_seconds_batch": 15.530873834970407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568047285079956, "target_z": 4.218749668259654, "ce": 5.959072113037109, "genuineness": 3.9801597595214844, "identity": 0.6687922477722168, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/37595508bf581342e3a2279047da35ce78c7819b198c55438755295bffe542a5.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "37e035b17df1ecde4c99ba594ce5a9f55808f433e0602dfc5ef8d63868f4fb45", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c906accea0f3d5aae483489f7653286c7ac87cab5d0bf0d0c630fa19ae9e89ec", "generation_seconds_batch": 15.683879307005554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08959884941577911, "target_z": -0.4277343813292566, "ce": 6.149696350097656, "genuineness": 4.7081804275512695, "identity": 0.6813443899154663, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/37e035b17df1ecde4c99ba594ce5a9f55808f433e0602dfc5ef8d63868f4fb45.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": "37e6ccb47eec08dcfac5f3df779e93606c29841b791163f630814519c1d5bd0f", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5593c27bbf70069ab73dbc68788a8c67b57fbae0cc3a5dccef6020480b335e3", "generation_seconds_batch": 18.94644432491623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.401228427886963, "target_z": 2.0312499685719674, "ce": 5.808161735534668, "genuineness": 4.250029563903809, "identity": 0.20121392607688904, "hypothesis": "De weg gaat linksaf naar de brug. Lop verder tot aan het kleine houtenhek. Ik wacht aan de andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/37e6ccb47eec08dcfac5f3df779e93606c29841b791163f630814519c1d5bd0f.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "37f0513d2f1fdc521b5e3b4044e9059eb5682d151dd24122942fea7fc922d27b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5345b85b2bb891246f6549aba495300767a9d16663c7cd17d1ec7f517f24f147", "generation_seconds_batch": 20.652124678948894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8346184492111206, "target_z": 2.843750157140164, "ce": 5.216432571411133, "genuineness": 4.557555198669434, "identity": 0.5967153310775757, "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/emo_Confusion/37f0513d2f1fdc521b5e3b4044e9059eb5682d151dd24122942fea7fc922d27b.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "3816eef5e7f6bb11c858a422ffb1ef19067f8bd1ef5619b5f11343004ab7eabb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "75a59b0f743bfa0e58132e198311a4e4a0cd50dda0b6b1346a2e21ac04361c0f", "generation_seconds_batch": 20.593501407071017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17398487031459808, "target_z": -0.26953125130950134, "ce": 6.252951145172119, "genuineness": 4.507347106933594, "identity": 0.5091870427131653, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3816eef5e7f6bb11c858a422ffb1ef19067f8bd1ef5619b5f11343004ab7eabb.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "382146a33217689bdb107a8e6e38274233ffee56a91af6ac5ce3f06788eb3b66", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29fa42bcc0b870a0020fb07f516d6ebd897fb3e33597795879a9d922a442f2ba", "generation_seconds_batch": 23.829392445040867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11772751808166504, "target_z": -0.3750000139680146, "ce": 5.895530700683594, "genuineness": 4.049196243286133, "identity": 0.2970946729183197, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/382146a33217689bdb107a8e6e38274233ffee56a91af6ac5ce3f06788eb3b66.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "3829bccb27f094a0aa174fbd34aa6e3dff11890b11ec8da994c079cc217ef3d4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b51aa9feea3bc8aaaab6cdaff69037fd39fca61e7d5a82d2193296e9498b7bcd", "generation_seconds_batch": 22.169447844964452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 6.062326908111572, "genuineness": 4.570107460021973, "identity": 0.6488212943077087, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3829bccb27f094a0aa174fbd34aa6e3dff11890b11ec8da994c079cc217ef3d4.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "387bbba5dec340605e03edfc59de67b7d91fb1cf72666ff53ca986a9d5e08dc7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "667b4321858d24e0a9ba44626fc9611df5bdede29b810263e692d23b05f6feb3", "generation_seconds_batch": 20.703628365998156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.151195764541626, "target_z": 1.5624999371439343, "ce": 5.617537498474121, "genuineness": 4.356721878051758, "identity": 0.49904006719589233, "hypothesis": "I f found a little box. It was beside the old window. We can look inside together.", "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/emo_Confusion/387bbba5dec340605e03edfc59de67b7d91fb1cf72666ff53ca986a9d5e08dc7.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "391e80896b27468b693f9bfac3047c6a68e1298f04143d7d0b7140f961b77eaa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "02ff88d1ce94de729230dc81c735ebca0c1cd46f776adcae5cecc0028d5e449a", "generation_seconds_batch": 23.959280307986774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20315533876419067, "target_z": -0.21484376440451505, "ce": 5.871703147888184, "genuineness": 4.432034492492676, "identity": 0.5223308801651001, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/391e80896b27468b693f9bfac3047c6a68e1298f04143d7d0b7140f961b77eaa.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "395910a1507fce452a9d70259dec4d77444e08f70388171da7d1518552cfd5c4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f45f4276a1eed624585120281c468a632329f60241165cc6aa43ef6c228b7e57", "generation_seconds_batch": 24.067310199025087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019798070192337036, "target_z": -0.5585937539285041, "ce": 6.070269584655762, "genuineness": 4.3504462242126465, "identity": 0.7487989068031311, "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/emo_Confusion/395910a1507fce452a9d70259dec4d77444e08f70388171da7d1518552cfd5c4.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "39ad50ed08ead1c01a7dd1449e72d5194e7e6e314d8e73083ef61b9fda170837", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "eb31651dc1138f51932b7e6b55f1de0199a6629360d6a86c98a6494951c22212", "generation_seconds_batch": 23.383483270998113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.984637975692749, "target_z": 3.125000041904044, "ce": 5.597680568695068, "genuineness": 4.356721878051758, "identity": 0.5664979219436646, "hypothesis": "He encontrado, la cajita estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/39ad50ed08ead1c01a7dd1449e72d5194e7e6e314d8e73083ef61b9fda170837.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": "39cdfe50dfe1a1d3c7ccfe4ef6658d82bd213bfb907ede09f514d64cbbc67b22", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "330548c463c2bca3636b65a6c3e7b817d1946a33adb24efa2728ca686c4ce5a5", "generation_seconds_batch": 15.908283584052697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17190127074718475, "target_z": -0.27343749039698995, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.6290304064750671, "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/emo_Confusion/39cdfe50dfe1a1d3c7ccfe4ef6658d82bd213bfb907ede09f514d64cbbc67b22.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": "3a9972313ccc04365dccb31209942e90641bef77108b78e30a081c2cf70812a5", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f498a670f10cb4a18ce197ce32692c77219122daf15301c21a201728c68b6f66", "generation_seconds_batch": 22.031887800083496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5448666214942932, "target_z": 0.42578123384948313, "ce": 5.959072113037109, "genuineness": 4.193545341491699, "identity": 0.3187214136123657, "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/emo_Confusion/3a9972313ccc04365dccb31209942e90641bef77108b78e30a081c2cf70812a5.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "a7eb1e4ad63564a0110d624c0404d8feee5f44f439d561d3195c782cf58f37f1", "generation_seconds_batch": 14.94309678603895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16148324310779572, "target_z": -0.29296874170649134, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.587454080581665, "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/emo_Confusion/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "3ca8cf44af0a82a28ceeb2a7eee6e2eab40d30b9c04c300c6c0c091eece1b062", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3044cd16a3cf188f086aaaafa35d3bcbc90d3292eaae33bc5ef8a276c3257c51", "generation_seconds_batch": 20.353285230929032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3138468861579895, "target_z": -0.007324212257055719, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.701383650302887, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3ca8cf44af0a82a28ceeb2a7eee6e2eab40d30b9c04c300c6c0c091eece1b062.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "3d2010a4d3adb4b06ffc4598a4c807aa06ccb8bd3d84c9593ee93854c0cc7b4f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b3075cc05b78206c3f4615ca37d25063ea465bd6a1e717a9317e05e3838f3258", "generation_seconds_batch": 21.461706755915657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5104871392250061, "target_z": 0.3613281212897461, "ce": 5.8875885009765625, "genuineness": 4.770940780639648, "identity": 0.7364757061004639, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3d2010a4d3adb4b06ffc4598a4c807aa06ccb8bd3d84c9593ee93854c0cc7b4f.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "3d2fcdf45f6a4fa554174c0762e0c416c13b63513f5ee3081641775d71f0de1f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2354dda70bfb5cd89d9982e4c733666f4fdd258563d8f42f936ab61e80f3c65", "generation_seconds_batch": 21.143578093964607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5281977653503418, "target_z": 0.3945312094054576, "ce": 6.022613525390625, "genuineness": 4.432034492492676, "identity": 0.5633542537689209, "hypothesis": "The door is sill open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3d2fcdf45f6a4fa554174c0762e0c416c13b63513f5ee3081641775d71f0de1f.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "3d34e5adcf77f140e588a325c92dd02458ce089dd9b84218fe2de03382122320", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0a1056c538ec307ad74111ac44b691adede03ab3fa9ad1f571ad94f9b85db7e7", "generation_seconds_batch": 21.690407857997343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8636583089828491, "target_z": 1.0234376021411067, "ce": 5.593709468841553, "genuineness": 4.532451152801514, "identity": 0.5852092504501343, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3d34e5adcf77f140e588a325c92dd02458ce089dd9b84218fe2de03382122320.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "3d5d4294381e63e2568429d0b67e0ee92ccd7d9205bb5b4cf206dda7eb3d0145", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "715ea76eaaff784115bf7e69ac07df9f74fc39b31bf9f705e664ea3fe0d36d90", "generation_seconds_batch": 22.31962552596815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07397180795669556, "target_z": -0.45703125829350866, "ce": 6.0305562019348145, "genuineness": 3.9299514293670654, "identity": 0.23230373859405518, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3d5d4294381e63e2568429d0b67e0ee92ccd7d9205bb5b4cf206dda7eb3d0145.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "3d6ba361e3cfaf6eebb8a3a55510f9041758cb0d441685a3c26b29e0cbc6fb74", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f9b0d1487965c4e91dbfcb02d7db7b7210397eb44373017b18bf75f5e594aa84", "generation_seconds_batch": 25.8729525039671, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.759477972984314, "target_z": 0.8281249773019763, "ce": 5.895530700683594, "genuineness": 4.6203155517578125, "identity": 0.5926017761230469, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3d6ba361e3cfaf6eebb8a3a55510f9041758cb0d441685a3c26b29e0cbc6fb74.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "3dbbab5bca5f2fe69e53c2ba7eed9db0645cf42a160a463c02019e6963706bc6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "73d5bf05067ec2e16f6ffe9148ac0a26d94a9981c4691386fcac4b87b1a4430a", "generation_seconds_batch": 21.46161303005647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5429136753082275, "target_z": 2.296875120474126, "ce": 5.871703147888184, "genuineness": 4.344170093536377, "identity": 0.46888405084609985, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3dbbab5bca5f2fe69e53c2ba7eed9db0645cf42a160a463c02019e6963706bc6.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "3e06223c831db966dbea5e6761cddcd64fd4fca62d32422cec8e733c34de786a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "44ec8dff8d3a48ccd1f808313a995912cf191e5fb4ca294d8be6b1d3ef42bf2b", "generation_seconds_batch": 19.256980390055105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4345661401748657, "target_z": 2.0937500174600183, "ce": 5.855817794799805, "genuineness": 4.507347106933594, "identity": 0.5055774450302124, "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/emo_Confusion/3e06223c831db966dbea5e6761cddcd64fd4fca62d32422cec8e733c34de786a.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "3ec64224a770bba2d5ad193eda042eb15afc1631e4a1cd3f47332e7c9a8741f3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c1fa48f36ef37a355160cd7ea5646a84d38b74f19c8251f215c98911513b2a03", "generation_seconds_batch": 20.995494131930172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08230623602867126, "target_z": -0.4414062460714959, "ce": 5.998785495758057, "genuineness": 4.758388519287109, "identity": 0.6640491485595703, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3ec64224a770bba2d5ad193eda042eb15afc1631e4a1cd3f47332e7c9a8741f3.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "3ef49100f85bb3d91dcb94fe0cfb53c80da55d93cf48e770015a0db49f0b6e39", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c1a8673f54319473f7376e20120915ced83dc464b316a61d75f881e6373aaf93", "generation_seconds_batch": 20.315113247954287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3161909282207489, "target_z": -0.0029297072516456277, "ce": 5.760505676269531, "genuineness": 4.68307638168335, "identity": 0.6694703698158264, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/3ef49100f85bb3d91dcb94fe0cfb53c80da55d93cf48e770015a0db49f0b6e39.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "3fa3b712d64842acde17531c2b6fc91d7e963605ff05bc7c25fef0c0e7bc69f7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5bcc993739091b78c9eb8b5d6aca47b62b93cb148a58c2fc28ea332f171a6be3", "generation_seconds_batch": 21.201648335088976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4595694541931152, "target_z": 2.140625109998115, "ce": 5.784333229064941, "genuineness": 4.607763767242432, "identity": 0.441201388835907, "hypothesis": "Ik 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/emo_Confusion/3fa3b712d64842acde17531c2b6fc91d7e963605ff05bc7c25fef0c0e7bc69f7.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "3fdde86cdc0d649aae60be902907b04a12058daf7dc10c9a3e28e221378f68ee", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4f892db00e8c46edbb7a932e35796c89018e0f90690f203ad3ce2288074b815e", "generation_seconds_batch": 19.628857400035486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.917962670326233, "target_z": 3.000000167616175, "ce": 5.2045183181762695, "genuineness": 4.18726921081543, "identity": 0.4689197540283203, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás linda.", "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/emo_Confusion/3fdde86cdc0d649aae60be902907b04a12058daf7dc10c9a3e28e221378f68ee.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "40c6a16112cbaed2a8f8e4b05be62ca11b07bd97ab246e423c94f79704b87e18", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d9f29833a6ca2bd182228d07638559fd659d49ebb51bdaa26674c50200ec5e01", "generation_seconds_batch": 19.373244078014977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3845596313476562, "target_z": 2.0000000558720585, "ce": 5.274016857147217, "genuineness": 4.607763767242432, "identity": 0.7699909210205078, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/40c6a16112cbaed2a8f8e4b05be62ca11b07bd97ab246e423c94f79704b87e18.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "40e4a58cd1c3193b7575bdf7875000434d5f4d69f3c86472380477ab269e67e6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e031d8380addba08625b4777a8b04c58a322dd25dc135293620a64574913f20f", "generation_seconds_batch": 16.276890425011516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08959884941577911, "target_z": -0.4277343813292566, "ce": 6.229123115539551, "genuineness": 4.595211505889893, "identity": 0.49201956391334534, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/40e4a58cd1c3193b7575bdf7875000434d5f4d69f3c86472380477ab269e67e6.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": "40e573863be8e8e0cd699a731d08e39d55d12199c02661ffcb54ed5111efdfc3", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8efbbcd6665fda3224a187d964f14679c0d276056c557cacbb21ba294c720989", "generation_seconds_batch": 20.784826064947993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846118211746216, "target_z": 2.749999972063971, "ce": 5.879645347595215, "genuineness": 4.356721878051758, "identity": 0.5285369157791138, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto al estantería alta. La colocaremos después de comer.", "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/emo_Confusion/40e573863be8e8e0cd699a731d08e39d55d12199c02661ffcb54ed5111efdfc3.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "416a4371036f222e5b28045a7312130ed13182abad302571ad5f528629da6f12", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a429ecccf0114a4f5fe8eb92ea49e3c2ae5adfb2944f338d756edd9fb701c26b", "generation_seconds_batch": 28.631439635995775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4323519468307495, "target_z": 0.21484376440451505, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.5213844776153564, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/416a4371036f222e5b28045a7312130ed13182abad302571ad5f528629da6f12.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "41c80a91abd792704d0d20ea13abf7d464cd032d57c93f7864bad56796456e52", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a325746f6e5494ea80132244a3d6cb032b4c4977e1f705f6db7cfa49ef91acc6", "generation_seconds_batch": 20.842595033929683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10626769065856934, "target_z": -0.39648438482126025, "ce": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.32305192947387695, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/41c80a91abd792704d0d20ea13abf7d464cd032d57c93f7864bad56796456e52.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "41cd5600dbcb17444adf77a9b141bcda2f216b86aed2e84df380615c77c6ae52", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3f3e12e4984de5c95cb3b5dd97162ea4db8d65d669abbba8678c7b916f1693b7", "generation_seconds_batch": 21.024782811990008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12085293233394623, "target_z": -0.3691406274007525, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.668968677520752, "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/emo_Confusion/41cd5600dbcb17444adf77a9b141bcda2f216b86aed2e84df380615c77c6ae52.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "41ed50d03be2da0ed78964b89cb39613eb5219714180d9fae205a0b76fda5f27", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3279211767cccdb95af05f74aaba1a94c059383880bd2d9d059296bc18ab8d70", "generation_seconds_batch": 15.999598938971758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00103798508644104, "target_z": -0.5976562565475069, "ce": 5.700935363769531, "genuineness": 4.143336772918701, "identity": 0.6365828514099121, "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/emo_Confusion/41ed50d03be2da0ed78964b89cb39613eb5219714180d9fae205a0b76fda5f27.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "42cb0ba3ec322657dfac1e070081ed1d31f257259a55945cc924fbe10a8892fa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "afbb353418a6832ae6e80c63352d236ef44df6e36953f67c72f9b8bba34f08fa", "generation_seconds_batch": 15.70324643899221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.117925643920898, "genuineness": 4.6705241203308105, "identity": 0.7761802673339844, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/42cb0ba3ec322657dfac1e070081ed1d31f257259a55945cc924fbe10a8892fa.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "43357aad723efc96b3ef646bd38f3d463f4e4066b1279a7af09ea2bb38bcfb6b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "db1bb50d51f7941d2c485f252008f38add07fddfddcfc5dc205dc72ee83a7126", "generation_seconds_batch": 19.625613541924395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.45422980189323425, "target_z": 0.2558593865672621, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.5635337829589844, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/43357aad723efc96b3ef646bd38f3d463f4e4066b1279a7af09ea2bb38bcfb6b.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "4345b508d663d503a16ad03ca793ebf0cf4cc2b2934f940b5abbda46202312be", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6defd869438c264e31b053e642bcde5dd3867524bfd0f0eb2655d7d46a2312f", "generation_seconds_batch": 15.608246298972517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9679691791534424, "target_z": 3.093750129204135, "ce": 5.637393951416016, "genuineness": 4.469690799713135, "identity": 0.49392902851104736, "hypothesis": "la puerta sigue abierta. Ah he dejado tu abrigo sobre la sobre la silla. Dime cuándo estás listo.", "wer": 0.2, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4345b508d663d503a16ad03ca793ebf0cf4cc2b2934f940b5abbda46202312be.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "b58cd0f374e62ee6d5dfe48dd5b1bc216520711699cccb3b185b06a7ed200733", "generation_seconds_batch": 16.312376318965107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08647345006465912, "target_z": -0.4335937399604895, "ce": 6.062326908111572, "genuineness": 4.570107460021973, "identity": 0.5300495028495789, "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/emo_Confusion/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "43777d477bfa182ae57e2fe09352657cde9070aefb3ad7e2ad4a57f2795bf5bf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1cd8dceacd629c654b5d79ac580ac424653ec0dee6aeef30e37f3579b78a9cc7", "generation_seconds_batch": 15.492544471053407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.442900538444519, "target_z": 2.1093749738099725, "ce": 6.038498878479004, "genuineness": 4.356721878051758, "identity": 0.3792923092842102, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/43777d477bfa182ae57e2fe09352657cde9070aefb3ad7e2ad4a57f2795bf5bf.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": "43951150879957774fa874d3474c773c3caf22400ac34c923b49ad293300e167", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "fe9ba51e7ef7a16ed7a3c930caaa0ffd445e9d5e1ff9349d98fe658942b2c9a7", "generation_seconds_batch": 16.898630206938833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3595563173294067, "target_z": 1.9531249633339618, "ce": 5.895530700683594, "genuineness": 4.381826400756836, "identity": 0.48622560501098633, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/43951150879957774fa874d3474c773c3caf22400ac34c923b49ad293300e167.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a2c96bd91063f13a2f0606ec0ebd2f30de07ffc625dab219a3d01d7ea4b84240", "generation_seconds_batch": 15.091096078976989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.372091293334961, "genuineness": 4.570107460021973, "identity": 0.5863412618637085, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "449f93fe871b5074afe9ae2ef10a41f600a6fe10434418377028bbc844ec29e3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "af3bc4ce1a75f6237eb091dbfd43fe9e3fcfa76b317de8e3ab5e135391d130fe", "generation_seconds_batch": 22.009176153922454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21148976683616638, "target_z": -0.1992187521825023, "ce": 5.486483573913574, "genuineness": 4.582659721374512, "identity": 0.5610568523406982, "hypothesis": "J'ai trouvé la 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/emo_Confusion/449f93fe871b5074afe9ae2ef10a41f600a6fe10434418377028bbc844ec29e3.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "4503a4b3b7f9de5ebb01c7eb60a4b9acf82e1d007f2cdccf505466cf41d4ebe4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7b4e80cd6596facc7baf5cad891942a605d4c7a1517666a0a148430489c10ff0", "generation_seconds_batch": 22.51326278306078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08751524984836578, "target_z": -0.4316406204167452, "ce": 5.959072113037109, "genuineness": 4.557555198669434, "identity": 0.6232283115386963, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4503a4b3b7f9de5ebb01c7eb60a4b9acf82e1d007f2cdccf505466cf41d4ebe4.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "455d63d4e7587fb0d6dc8ffda475e206c0da6890ec88054d7101135d93c806ba", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9dc85bf7154bc45fd9a371e783f09323177c3ccd42da5ba3db3800903c4535f0", "generation_seconds_batch": 18.616849964950234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36776018142700195, "target_z": 0.09375001746001824, "ce": 5.974957466125488, "genuineness": 4.174716949462891, "identity": 0.34619683027267456, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/455d63d4e7587fb0d6dc8ffda475e206c0da6890ec88054d7101135d93c806ba.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "4fd9fe7ca88fee351c8e01598e72d3465be0dbe754447c6fcf1f4ef07edf2897", "generation_seconds_batch": 15.029856837936677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.005212843418121338, "target_z": -0.5859374834129827, "ce": 6.324435234069824, "genuineness": 4.093128204345703, "identity": 0.5647069215774536, "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/emo_Confusion/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "4606d214d3e02a4c492c184f31432e8a5f0491cd1bda628fb3d3740ccf05edb8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8eeb304112668f079029a08b0265799ccf31af335e68cfc05e0a899ddef6038", "generation_seconds_batch": 18.616849964950234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.431831032037735, "target_z": 0.21386717669661373, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.4254218339920044, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4606d214d3e02a4c492c184f31432e8a5f0491cd1bda628fb3d3740ccf05edb8.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "4677d7eeaa256facbc610d7a614a4d4d4ce4482664eff0182302a51fe0b43024", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cac1ada6453cb6b9f7d87846c4e75ef5988a24722f5311875d370c9f670047f9", "generation_seconds_batch": 15.782314590062015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08855704963207245, "target_z": -0.4296875008730009, "ce": 6.1020402908325195, "genuineness": 4.406930446624756, "identity": 0.6810062527656555, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4677d7eeaa256facbc610d7a614a4d4d4ce4482664eff0182302a51fe0b43024.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "4791cdad60bb83c78e6e558c054c22b367e338fbc89821a680274aa8c04139ef", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "eaa8568986f371729f25aadfc15c64a81c9fc509f6759ee42ae7253643cc627b", "generation_seconds_batch": 19.809968107030727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.5575162172317505, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4791cdad60bb83c78e6e558c054c22b367e338fbc89821a680274aa8c04139ef.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "47c1cdc9351047898af39f75018a5bdab96d82b9d6975ae301648df3f4b89757", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "19f5bc6db240e8a545120630afd5b50743f4d334a9d18f1ceeb1396e0f4aba6e", "generation_seconds_batch": 20.995494131930172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07397180795669556, "target_z": -0.45703125829350866, "ce": 6.284721851348877, "genuineness": 4.758388519287109, "identity": 0.6761002540588379, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/47c1cdc9351047898af39f75018a5bdab96d82b9d6975ae301648df3f4b89757.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "bcb4462ccee961749ddd962efe2c02658766ba96d03361d91c389f613bda5386", "generation_seconds_batch": 15.964063439052552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.7730001211166382, "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/emo_Confusion/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "48ad73905fd898853873faf529fdae8f693285edb3a5e91e3a059d383a5b25b5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a155c990087589ac347d047af8bc098a25bc16942fa1fa54bece6333ef0585f7", "generation_seconds_batch": 23.01439790194854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0928548574447632, "target_z": 1.45312501920602, "ce": 5.4388275146484375, "genuineness": 4.425758361816406, "identity": 0.827970027923584, "hypothesis": "De 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/emo_Confusion/48ad73905fd898853873faf529fdae8f693285edb3a5e91e3a059d383a5b25b5.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "4986321e4e2e25e89dc71608c821e56f413be483669269a51a3310b910a6b37d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "447716f60673ffa099c86f5f08776c39d35c722d72323fb6c2dd85a8a26c9b5f", "generation_seconds_batch": 21.268855940084904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05938658118247986, "target_z": -0.4843749877779872, "ce": 6.356205940246582, "genuineness": 4.532451152801514, "identity": 0.5294862985610962, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4986321e4e2e25e89dc71608c821e56f413be483669269a51a3310b910a6b37d.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "1f663438090c202777ecf0f1e933243689ed559ca175b738b05de9aeb0e2962f", "generation_seconds_batch": 15.384951777989045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.011463642120361328, "target_z": -0.5742187661505168, "ce": 6.1894097328186035, "genuineness": 4.243753433227539, "identity": 0.6955921649932861, "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/emo_Confusion/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "4a3d48c563b196b3ed6b89df1026e44e088fc8e6f3cfac082e3e9515df9acdf0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ab1f1ea04ca3e638a6dcaa6d01e9ee8018a3c2b2a830a71c465c58ab143922fa", "generation_seconds_batch": 20.015073664020747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.013539612293243408, "target_z": -0.6210937469444968, "ce": 6.427689552307129, "genuineness": 4.362998008728027, "identity": 0.4235800504684448, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4a3d48c563b196b3ed6b89df1026e44e088fc8e6f3cfac082e3e9515df9acdf0.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "4a68a26f1bd51e363456bf6a22b7c46945686fe4a9d24fa1b308e232e7308902", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "036a8cfbac70901bee4c4696ad7b97a2da18054c3b762dca46c409ab336d9361", "generation_seconds_batch": 19.962014107964933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1429920196533203, "target_z": 3.421875106506111, "ce": 5.967014789581299, "genuineness": 4.18099308013916, "identity": 0.4862903952598572, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4a68a26f1bd51e363456bf6a22b7c46945686fe4a9d24fa1b308e232e7308902.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "4ba6d58b8a8fc5dc540a2adab4bfda6137715dab1c22dfaa0a785f33e361a82c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4a62dbb83ec06796cfd174b4c48a9af8763451b33f170774d95f68a118e4c078", "generation_seconds_batch": 20.78185544698499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2345399856567383, "target_z": 1.7187499476199453, "ce": 6.054384231567383, "genuineness": 4.381826400756836, "identity": 0.6924135684967041, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4ba6d58b8a8fc5dc540a2adab4bfda6137715dab1c22dfaa0a785f33e361a82c.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "4bb6ff4648de37b32cbc15797d2eeb544c5b8568f126e0b7366b5a51e5528b82", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "dd765a708bcc59940cee561531e24b5831d212c82345a724ed87ebbe641159c9", "generation_seconds_batch": 22.191772917984053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.200550839304924, "target_z": -0.2197265632638758, "ce": 6.324435234069824, "genuineness": 4.519899368286133, "identity": 0.38571029901504517, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4bb6ff4648de37b32cbc15797d2eeb544c5b8568f126e0b7366b5a51e5528b82.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "4bd6e997595b83501f2a1b5359f21794e728f5e60efa5dc56933398368c8c662", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a02b422cccf4c8e641a96668550bdf92f5f6e94c25eed762f0d41b3207558f74", "generation_seconds_batch": 21.45652892801445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1073095053434372, "target_z": -0.39453123734148676, "ce": 5.538110733032227, "genuineness": 4.18726921081543, "identity": 0.4964858591556549, "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/emo_Confusion/4bd6e997595b83501f2a1b5359f21794e728f5e60efa5dc56933398368c8c662.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "4cb29ad4b42faa94d574619f53135e5e9dce00feef1394198a64bf4146003260", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "14bee6cee28b965a26babcdc2bd007b619fb2d3d1f312ef9e1b45e7302439083", "generation_seconds_batch": 22.1693549240008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1720318794250488, "target_z": 1.6015625515070537, "ce": 5.831989288330078, "genuineness": 4.6203155517578125, "identity": 0.7172269821166992, "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/emo_Confusion/4cb29ad4b42faa94d574619f53135e5e9dce00feef1394198a64bf4146003260.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "4cd3d0e140d53a43a486ea41cfcd947cbcd2f99d1888d0e070b56430aec29246", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "66c38b6296fbde35c45ee531197a884568f1bf7a0681237850d84f0aa58b7a5e", "generation_seconds_batch": 20.169644415960647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2553761005401611, "target_z": 1.7578125619830647, "ce": 5.534139156341553, "genuineness": 4.13078498840332, "identity": 0.26201626658439636, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4cd3d0e140d53a43a486ea41cfcd947cbcd2f99d1888d0e070b56430aec29246.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "4d63745e7abfe0ae580fc7ac40ab459b6ba1971c35cb60cb2c7b03582b66cf1c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "22480d4f8d24d417325f9341669ebfb8fdd0ccb8757d87c1764efeb0889918d0", "generation_seconds_batch": 21.69620088103693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07813902199268341, "target_z": -0.4492187521825023, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.6165735125541687, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4d63745e7abfe0ae580fc7ac40ab459b6ba1971c35cb60cb2c7b03582b66cf1c.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "4e48153dfb371e20369d1bf11ecb9ca95ccc3362780150b06582e0f08b45a44d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c9d9062146f844c74987b6e5423889e6ecc79d820a8590fdea9fb3a1e8c28bc2", "generation_seconds_batch": 20.52017790300306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2213868945837021, "target_z": -0.18066406064487306, "ce": 6.141753673553467, "genuineness": 4.65797233581543, "identity": 0.737507700920105, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4e48153dfb371e20369d1bf11ecb9ca95ccc3362780150b06582e0f08b45a44d.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "4e8ccca6fcae8e0cff986e4a614fd3e50f1c8cc21316109852c139b109fa53a4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ac5889087c9ead057522aeb64ff59d830511150150745b687c1988f18028b744", "generation_seconds_batch": 20.3413369449554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.200550839304924, "target_z": -0.2197265632638758, "ce": 6.0305562019348145, "genuineness": 4.18726921081543, "identity": 0.48761534690856934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4e8ccca6fcae8e0cff986e4a614fd3e50f1c8cc21316109852c139b109fa53a4.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "4e998b23aa00fdaa1b2df31df5682d3e13de932715e57f0c78a6a31ac4fe9aca", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9c14150fbc20241fa6f23025a61caedf15d0556b098e40a425bb98410003276f", "generation_seconds_batch": 20.009339048992842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.334683656692505, "target_z": 3.7812499965079964, "ce": 5.34748649597168, "genuineness": 4.143336772918701, "identity": 0.11416205018758774, "hypothesis": "De deur staat nog open. Ik heb je inhails op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/4e998b23aa00fdaa1b2df31df5682d3e13de932715e57f0c78a6a31ac4fe9aca.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": "4f4d27962d0fee203b2da0afa841db4df98fd79fadd891b5d4b81093b39a7892", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c67edcd8f7d58a270da052da34552ef3356ca3f0d5b239c5b44ae1cbb25522b", "generation_seconds_batch": 21.921301924041472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6198763847351074, "target_z": 0.566406176231423, "ce": 5.4944257736206055, "genuineness": 4.758388519287109, "identity": 0.6031541228294373, "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/emo_Confusion/4f4d27962d0fee203b2da0afa841db4df98fd79fadd891b5d4b81093b39a7892.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "5042e961f5df82f400d36af57d3cbae60d8a5a0cc1edfe0ca8f081c86f547ab3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9ed6f459a0425c5eb016f43344762b4bba4c37b7cdcb07dbbcaaa18650920fff", "generation_seconds_batch": 20.968235336011276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07813902199268341, "target_z": -0.4492187521825023, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.6046048402786255, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5042e961f5df82f400d36af57d3cbae60d8a5a0cc1edfe0ca8f081c86f547ab3.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "505549f82a3acef3edd4bcd6f999ecacb9398ae875ee330f628d8fe5b7a072ac", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "faacb15702da0b229f06bd22be2f0c447e68a63c5ae73935b38c6819ec702b99", "generation_seconds_batch": 13.993220232077874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11251851916313171, "target_z": -0.3847656116867361, "ce": 6.221180438995361, "genuineness": 4.7332844734191895, "identity": 0.3118530809879303, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/505549f82a3acef3edd4bcd6f999ecacb9398ae875ee330f628d8fe5b7a072ac.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "5060c1eb0669bda054eb308bbe7f8388eb8a8fcb4ee2783c3acb0729ce378385", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "095ce315ec0537abe197f88c2a67812e59ba0d1c3fee423d3f837d074c96fa82", "generation_seconds_batch": 21.268855940084904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009380042552947998, "target_z": -0.5781250052380055, "ce": 6.229123115539551, "genuineness": 4.21237325668335, "identity": 0.45459964871406555, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5060c1eb0669bda054eb308bbe7f8388eb8a8fcb4ee2783c3acb0729ce378385.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "5081c2c9cc0f57df88907102ca772711c4697663c0eb0f7d74823a0ed7ee05f7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1ed5bbbc73046030b3ae72e5acfef056b4c6244bb3f89461bc40509480a96f3b", "generation_seconds_batch": 21.4163426309824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7596086263656616, "target_z": 2.7031251030141075, "ce": 5.514282703399658, "genuineness": 4.400654315948486, "identity": 0.6203500032424927, "hypothesis": "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/emo_Confusion/5081c2c9cc0f57df88907102ca772711c4697663c0eb0f7d74823a0ed7ee05f7.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "50832466e7d4c57302fb0a4260ea5dd3197610f13ef3accb7c958bff1d5a5788", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c6c87196dc83c811f345fe5077a948751ecd7d4820ba4f403975e5ceadb01e84", "generation_seconds_batch": 20.391116392915137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26149630546569824, "target_z": -0.10546873472248404, "ce": 5.959072113037109, "genuineness": 4.6705241203308105, "identity": 0.7803138494491577, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/50832466e7d4c57302fb0a4260ea5dd3197610f13ef3accb7c958bff1d5a5788.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "511018eb1c10f30be0dca41ce5ddb01b7a446949daf0f6918325a583937b0de7", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6c67d4a240012acbf6f2e94376605ace2ad4b7a4d7ebdc23736a8b7155d7c8b8", "generation_seconds_batch": 15.683879307005554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00312158465385437, "target_z": -0.6015624956349954, "ce": 5.9431867599487305, "genuineness": 4.356721878051758, "identity": 0.5982241034507751, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/511018eb1c10f30be0dca41ce5ddb01b7a446949daf0f6918325a583937b0de7.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "513805494e6aa2c012ed85669dda8b1f5339c3f735cda71233f21c383f6fbbae", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ed3e9f6da3dae1cfb49234e02dceb4a5db56592488f0aeb66e808970e731772b", "generation_seconds_batch": 19.816080200020224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1302291601896286, "target_z": -0.3515624956349954, "ce": 6.117925643920898, "genuineness": 4.319066047668457, "identity": 0.2271624654531479, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/513805494e6aa2c012ed85669dda8b1f5339c3f735cda71233f21c383f6fbbae.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "51cf44491780b5382e3ac7b25fd633d27c38cc1867485be6bd7a85d3d571ce73", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "27055d869c936760ac5b32eb750ab77af8239b0f4ab3b31b23153c74f839f1b6", "generation_seconds_batch": 20.95043559395708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2052389532327652, "target_z": -0.21093749738099726, "ce": 6.205295085906982, "genuineness": 4.595211505889893, "identity": 0.41031414270401, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/51cf44491780b5382e3ac7b25fd633d27c38cc1867485be6bd7a85d3d571ce73.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "51d8f1ff448e3beb9bb70eafeeaf26d0cd57cda24b43ecbeafe50e76a5599208", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8c2da54be5b879f6ebd41d9f9f4c29273f00aa0d218b1453d6048bb9cdf02850", "generation_seconds_batch": 27.581000043894164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.056261152029037476, "target_z": -0.4902344022812785, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.43093645572662354, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/51d8f1ff448e3beb9bb70eafeeaf26d0cd57cda24b43ecbeafe50e76a5599208.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "51dab553ebba76458de4a0ccf1b631b5ae0c41f9976e52c7574b53f0ae7563ae", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3650e13f0224e8c1004288569df37efad713e02fdf2fd28508755d86e432009d", "generation_seconds_batch": 21.80175695195794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8678255081176758, "target_z": 1.0312500803160838, "ce": 5.661221981048584, "genuineness": 4.595211505889893, "identity": 0.6005963087081909, "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/emo_Confusion/51dab553ebba76458de4a0ccf1b631b5ae0c41f9976e52c7574b53f0ae7563ae.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "525bdd20a6b7ccc981a58cd8727529a420de9c1b7122e0726b72eee564396e5b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8266eca9e019150f13e02fcf1fcc22bfe2c0d80f5a15f9cbe5b279ca073064ec", "generation_seconds_batch": 15.009845591965131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06563737988471985, "target_z": -0.47265627051552145, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.37748491764068604, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/525bdd20a6b7ccc981a58cd8727529a420de9c1b7122e0726b72eee564396e5b.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": "527de73c69268bf289274a7017516acb1ce0b0c027a82994fe40b7b751660dec", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d0c6814315b1130b66509eab571d763e70f0b96391d5bb3196b5a7004eede30", "generation_seconds_batch": 16.898630206938833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2180018424987793, "target_z": 3.5625001606321676, "ce": 5.669164657592773, "genuineness": 3.9801597595214844, "identity": 0.5277713537216187, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/527de73c69268bf289274a7017516acb1ce0b0c027a82994fe40b7b751660dec.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "52809c20244922ba939a437766cfb01a9271011efba9f37ce98fd09c89e8bb8e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b0eafd31ddc7588d3712397c713f13a1b8084a438610b3d165ec82b5cbf0520e", "generation_seconds_batch": 20.91269912896678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.517910361289978, "target_z": 2.250000027936029, "ce": 5.363371849060059, "genuineness": 4.582659721374512, "identity": 0.6277192831039429, "hypothesis": "La porte est encore ouverte. Pousser ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/52809c20244922ba939a437766cfb01a9271011efba9f37ce98fd09c89e8bb8e.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "5290513160848706a60352d61f87a3e28b5a6f62bab96492657540182f642101", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2ed4ed0fbcdba73c604b97b7690a1980430a63e41c6e189eb9f82c9580ec7e91", "generation_seconds_batch": 21.996295294957235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293011426925659, "target_z": 3.7031247677817576, "ce": 5.677107334136963, "genuineness": 4.074300289154053, "identity": 0.14527562260627747, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5290513160848706a60352d61f87a3e28b5a6f62bab96492657540182f642101.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "5293939105901ec3ebdcd54c4e725a4cff5888d5d9a90abda2e6239e7aa05adf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "02756a3cc1a08db688a7037e89b3e200f5661e37909974cdc95432d34a40468b", "generation_seconds_batch": 15.600008339970373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0803532600402832, "target_z": 1.4296875846810884, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.6223584413528442, "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/emo_Confusion/5293939105901ec3ebdcd54c4e725a4cff5888d5d9a90abda2e6239e7aa05adf.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "52ac7b523e55d6a21f4504bbce4c772aaf71bd4b021b9adeec4c142afd6a70aa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "15585ccf770c004130d5492cc0c89adc781e970b05247d00b5b916a122dcac3a", "generation_seconds_batch": 14.863663148018532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11460211873054504, "target_z": -0.3808593725992475, "ce": 6.332377910614014, "genuineness": 4.545003414154053, "identity": 0.35127049684524536, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/52ac7b523e55d6a21f4504bbce4c772aaf71bd4b021b9adeec4c142afd6a70aa.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "53da401983bb18ddf555086accc6e81c3e9d983645739545222503d6f3231fef", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1412b8a77edabaf5cc0c07d0781590419805de4f26b5ddd99094766619febf13", "generation_seconds_batch": 21.57632663694676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8115681409835815, "target_z": 0.9257812897215415, "ce": 4.980137825012207, "genuineness": 4.3881025314331055, "identity": 0.7371513843536377, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/53da401983bb18ddf555086accc6e81c3e9d983645739545222503d6f3231fef.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "5446e301e7dfaca9a711786ee4df8dd993dfd240495b2cb1b405097ba7238dc7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b04922410a008b542ca28ba89b76e3fa145efc318cc3254a24b8850d644d977b", "generation_seconds_batch": 24.909868221031502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42662203311920166, "target_z": 0.20410157897789222, "ce": 5.681078910827637, "genuineness": 4.444586753845215, "identity": 0.6232866048812866, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam...", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5446e301e7dfaca9a711786ee4df8dd993dfd240495b2cb1b405097ba7238dc7.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "c3ced4ce3a470cafa3ea3378a33e75261929af44d124e7ad6d81fe6a85293f16", "generation_seconds_batch": 15.029856837936677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017714470624923706, "target_z": -0.5624999930159927, "ce": 6.213237762451172, "genuineness": 4.017816066741943, "identity": 0.6045719981193542, "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/emo_Confusion/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "54c163dce57cc6614a8c806321d13a499884f1558ed67b76c00051b93c04ccbc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a24d43c70b39a472dad6af388870932ebac5e60bb8e420549e5832a5bd20e053", "generation_seconds_batch": 20.732215017080307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8011500835418701, "target_z": 0.9062499825399818, "ce": 5.919358730316162, "genuineness": 4.507347106933594, "identity": 0.511199414730072, "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/emo_Confusion/54c163dce57cc6614a8c806321d13a499884f1558ed67b76c00051b93c04ccbc.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "55541baeba0147b481840600cc581d13a37ed2a805f584d489d6caec64e8baa1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f36a4350290e1bf3646764e1757c3c253838201decc5c740e8f36c303df602af", "generation_seconds_batch": 22.0632179339882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0720187425613403, "target_z": 1.4140624048429007, "ce": 5.673136234283447, "genuineness": 4.557555198669434, "identity": 0.6629879474639893, "hypothesis": "la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.47368421052631576, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/55541baeba0147b481840600cc581d13a37ed2a805f584d489d6caec64e8baa1.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": "55624720071dfae8d20226c72e23578ab5af1f4b241ef904fbf82a2051845d54", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75b67d27469db2bb8149800880c7a4fb730481d491c76e83e7a4b007d5192160", "generation_seconds_batch": 15.43189180700574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.592920184135437, "target_z": 2.3906250820620856, "ce": 5.019851207733154, "genuineness": 4.281409740447998, "identity": 0.4904107451438904, "hypothesis": "Escribir número en el sobre leerlo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "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/emo_Confusion/55624720071dfae8d20226c72e23578ab5af1f4b241ef904fbf82a2051845d54.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": "55812d68e058331e819eaf7bf121f11fb7b07c96cddf41b36f3267b79261d7af", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d12e1f746a2631ee94b587648b3b0db826c2bd760c7f5a8b8c577ca686db4248", "generation_seconds_batch": 14.019105428946204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5845857858657837, "target_z": 2.3750001257121314, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.376548171043396, "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/emo_Confusion/55812d68e058331e819eaf7bf121f11fb7b07c96cddf41b36f3267b79261d7af.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "5585cab98a4399dde7e6919d41a231b061974435ef1a18ed44deb22910558b32", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "bbd29f03d1efbbc83506bfd2e9095b16c13fba473e874015c4c47e871f08d8ad", "generation_seconds_batch": 21.398158353054896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.786564826965332, "target_z": 0.8789061971834449, "ce": 6.0305562019348145, "genuineness": 4.293961524963379, "identity": 0.615419864654541, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5585cab98a4399dde7e6919d41a231b061974435ef1a18ed44deb22910558b32.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "55991f76244725d78938eb8aa4f6b41adacfd8ffa907c2db0e7efad1ab232b17", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fea3bb8e80190fc7312be51f421c8d736f73c35eedc7ac3d6dd6f55e2e586f4a", "generation_seconds_batch": 21.746152767911553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.667930006980896, "target_z": 2.531250136188142, "ce": 5.403085231781006, "genuineness": 4.093128204345703, "identity": 0.3618859052658081, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/55991f76244725d78938eb8aa4f6b41adacfd8ffa907c2db0e7efad1ab232b17.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "564bf2aba5fdb981f29ee2863aeca70d0224795c4b94f4ba8e182345b5d5331c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "083c6c76876b50e160a1df924c826b22fc6666b86e17c875907d1cd4f9bc9775", "generation_seconds_batch": 20.62079343898222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8679560422897339, "target_z": 2.9062499825399817, "ce": 5.609594821929932, "genuineness": 4.300237655639648, "identity": 0.7639048099517822, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/564bf2aba5fdb981f29ee2863aeca70d0224795c4b94f4ba8e182345b5d5331c.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": "56524ff7a6d730377ae5d99f34f13dfce0d2c1a64917f57ec974b9ae2e2d60ed", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0259fc361d22defaf23451cf9cb77d5c6d59e5d6e34fc9affa49d224effcab1b", "generation_seconds_batch": 16.091904192930087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.376225233078003, "target_z": 1.984375099522104, "ce": 5.418970584869385, "genuineness": 4.507347106933594, "identity": 0.5720329880714417, "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/emo_Confusion/56524ff7a6d730377ae5d99f34f13dfce0d2c1a64917f57ec974b9ae2e2d60ed.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "56885343c691a8f73f098d814a270971ccd15fecdb7ed6d70d6e078981de27bc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a450dd3c97dcc06b8e9131422047593d15fd19578603884aa40adde7086f4095", "generation_seconds_batch": 19.32685847999528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 6.086154937744141, "genuineness": 4.256305694580078, "identity": 0.6728880405426025, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/56885343c691a8f73f098d814a270971ccd15fecdb7ed6d70d6e078981de27bc.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": "5695b90474f0f35e5870f4cb86441f04a2571a843fa6604ea953796fa2e02b12", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97859919ddfe7fe2e3f947833bef23a4ab0ae3f64b8092d7d62f1cf9c9fa950c", "generation_seconds_batch": 16.091904192930087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4513654708862305, "target_z": 3.999999832383825, "ce": 5.700935363769531, "genuineness": 4.268857479095459, "identity": 0.7014288306236267, "hypothesis": "Las llaves están en la mesa del cocina. Miren el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.17391304347826086, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5695b90474f0f35e5870f4cb86441f04a2571a843fa6604ea953796fa2e02b12.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "56dd09133fb80fa582759ee2c24589faeef4feb54454d86fa3fccde50fe2e0a3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cc3adf54504e8841eb39b7cd188eb24c49480185374b4beb4f6df21c0ee8adbc", "generation_seconds_batch": 22.168654451030307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.90962815284729, "target_z": 2.9843749877779873, "ce": 5.720792293548584, "genuineness": 4.369274139404297, "identity": 0.6186962127685547, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/56dd09133fb80fa582759ee2c24589faeef4feb54454d86fa3fccde50fe2e0a3.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "57e9428b9807df34101f3a04ad07dd9b3a020e11f0e62cfedb15a178764cf252", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9ae94215e49b1022fcf24ef15bc17d927b1234b55452d66d9a8f9111c2ce3274", "generation_seconds_batch": 15.671040626009926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9720057249069214, "target_z": 1.226562481666981, "ce": 5.77639102935791, "genuineness": 4.545003414154053, "identity": 0.6855116486549377, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/57e9428b9807df34101f3a04ad07dd9b3a020e11f0e62cfedb15a178764cf252.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "57fe448970974f7f0607e9a2df07e6440e8a0d2a6ab25e53d1e9b1e1204c08e9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b10ec709d2891676fefacf46156c13782e7d8737f0600754a397fef842eaaf45", "generation_seconds_batch": 22.26091180893127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08647345006465912, "target_z": -0.4335937399604895, "ce": 6.252951145172119, "genuineness": 4.7834930419921875, "identity": 0.753663957118988, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/57fe448970974f7f0607e9a2df07e6440e8a0d2a6ab25e53d1e9b1e1204c08e9.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "58229d821584b2055e040340ae29aaaf94598ccb5331399dc3bb6286589f8518", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5189acc8dc5fc102dfde57c794d453d606f1f3b7fe4833ef5657732b7c55d4ff", "generation_seconds_batch": 21.073654906940646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2513387203216553, "target_z": -0.12451172011406393, "ce": 5.649308204650879, "genuineness": 4.432034492492676, "identity": 0.6036814451217651, "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/emo_Confusion/58229d821584b2055e040340ae29aaaf94598ccb5331399dc3bb6286589f8518.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "58fc473f1b0499dea528cdfe626ae587b45be702ba0a6e2cf6c0446957f6b2b8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8c84e09553d57891348162994c0460cce797dedbae1f2701873685a44207ed97", "generation_seconds_batch": 22.078123525949195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1679952144622803, "target_z": 3.4687499755559745, "ce": 6.046441555023193, "genuineness": 4.331617832183838, "identity": 0.7258193492889404, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/58fc473f1b0499dea528cdfe626ae587b45be702ba0a6e2cf6c0446957f6b2b8.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": "597676c0a9c7341468e577382833da4e5ddc61772e9727ed18411955328773e5", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2b473fd5ea36f1253b212e69015900978351e96e828d921743168bfe7487290", "generation_seconds_batch": 20.982111165998504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.151195764541626, "target_z": 1.5624999371439343, "ce": 5.1767191886901855, "genuineness": 4.507347106933594, "identity": 0.7948992252349854, "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/emo_Confusion/597676c0a9c7341468e577382833da4e5ddc61772e9727ed18411955328773e5.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "59786ea7ba1bafcee4bd37a72a11eb0dda5439107505ccc964c533d3a2c37e5b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32e0cee1bf7b071b25b0f2d4e7e23decf6973ebce72aaf182d3de5d9a6ef4f99", "generation_seconds_batch": 15.009845591965131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05938658118247986, "target_z": -0.4843749877779872, "ce": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.5685003995895386, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/59786ea7ba1bafcee4bd37a72a11eb0dda5439107505ccc964c533d3a2c37e5b.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "59e8599d846d638d4b69bc17d0980c41f885f5b423c49c5ec420e05224b7f046", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d2d16dc7fef7c5a154359cd517c6c14598cf4e3c6782a47125602938f8fac811", "generation_seconds_batch": 20.5009416899411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.141753673553467, "genuineness": 4.3127899169921875, "identity": 0.5852743983268738, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/59e8599d846d638d4b69bc17d0980c41f885f5b423c49c5ec420e05224b7f046.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": "5a26e5efcba73ed9d39e5fa173d1df6905bfa257072a344d8726c59180e1082a", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f280cb49a30343cb205f8b95e94914cf1790c1d7a473e387f8781177497a948", "generation_seconds_batch": 15.314731306047179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2512089014053345, "target_z": 1.7500000838080876, "ce": 5.728734970092773, "genuineness": 4.632867813110352, "identity": 0.7201378345489502, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuánto podemos hablar.", "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/emo_Confusion/5a26e5efcba73ed9d39e5fa173d1df6905bfa257072a344d8726c59180e1082a.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": "5a46879265b9b8fee638d5ad995752acc9db98fadd8125467871656b4b507b5c", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf6c36b58c7a3878e5c69977b600e541de3a6d58d8d7ab10340f68e47ef026c6", "generation_seconds_batch": 20.30674255208578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5657026767730713, "target_z": 0.46484373646848587, "ce": 5.561938762664795, "genuineness": 4.13078498840332, "identity": 0.5905617475509644, "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/emo_Confusion/5a46879265b9b8fee638d5ad995752acc9db98fadd8125467871656b4b507b5c.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "5a70aa41d1db9ad3aa520218c25896dd69e8220aeec66f4d3e2ea634762d1ba9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bcc44eb51deecf0cdbf439d7dfa406bee0f943abde3f8cf942b1581924cef983", "generation_seconds_batch": 20.83052072895225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2281586080789566, "target_z": -0.16796875567450592, "ce": 6.062326908111572, "genuineness": 4.469690799713135, "identity": 0.4360797107219696, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5a70aa41d1db9ad3aa520218c25896dd69e8220aeec66f4d3e2ea634762d1ba9.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "5ab3353b47825a4d7a52e1ba9a6c4ac666f22bd1886e6f3516c74c659d6d8fcf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03572908c7c5ea8911308a31b4ff0c23553d98814d61d2cbec5757866739965b", "generation_seconds_batch": 20.762236740090884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15835782885551453, "target_z": -0.29882812827375344, "ce": 6.022613525390625, "genuineness": 4.494794845581055, "identity": 0.20136021077632904, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5ab3353b47825a4d7a52e1ba9a6c4ac666f22bd1886e6f3516c74c659d6d8fcf.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "5ad7a84b13be551a1c412ed7338d27e3c06ee8b5b7fc13d273136c64e1c32bab", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7e8d344154fb6447da3cbb643b95409858e4eb72ef310410c18b0b19345e65b5", "generation_seconds_batch": 21.51424367609434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07397180795669556, "target_z": -0.45703125829350866, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.5488568544387817, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5ad7a84b13be551a1c412ed7338d27e3c06ee8b5b7fc13d273136c64e1c32bab.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "5b1b77e3c1ce8ecb0b9f94dfbdb28561e12deccb670bdffa3b4616eeda8a6cb4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "41d17457ea697b98c5983719cf90ba5593c9e6666e721e1d0535269124f8eeb1", "generation_seconds_batch": 20.643940870068036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.056261152029037476, "target_z": -0.4902344022812785, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.4981518089771271, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5b1b77e3c1ce8ecb0b9f94dfbdb28561e12deccb670bdffa3b4616eeda8a6cb4.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": "5b44414f141b0979ca1ed2d5928ab7f786f2753a1e6f027df661a70e4203d6f0", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "096ea08773b6cf6b7eb9f3c729725be892f355d3cd008c83f3a4bb4b059a22e3", "generation_seconds_batch": 16.207539557944983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2470417022705078, "target_z": 1.7421876056331103, "ce": 5.557967185974121, "genuineness": 4.68307638168335, "identity": 0.8145143985748291, "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/emo_Confusion/5b44414f141b0979ca1ed2d5928ab7f786f2753a1e6f027df661a70e4203d6f0.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": "5b5aa0cc116eb6ddf6a281dad901b016c2982ad293765d771f6de9d258a8643e", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfb20578a88e2125bb624886240ffd06f73f20d7a82fa6a8a606545f0f86a9f2", "generation_seconds_batch": 25.94291638804134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.313717007637024, "target_z": 1.867187479920979, "ce": 5.879645347595215, "genuineness": 4.13078498840332, "identity": 0.17327946424484253, "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/emo_Confusion/5b5aa0cc116eb6ddf6a281dad901b016c2982ad293765d771f6de9d258a8643e.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "5b7c7cc250672740d5854e9c56f3dd3111d030237993a3a388b243c6f94e278b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6b4ea0244525956c3c83f120cf5cc91686d871359be6607e4ddc2ef802757d27", "generation_seconds_batch": 21.999544403981417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10001687705516815, "target_z": -0.4082031300197552, "ce": 6.070269584655762, "genuineness": 4.532451152801514, "identity": 0.8420105576515198, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5b7c7cc250672740d5854e9c56f3dd3111d030237993a3a388b243c6f94e278b.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "5c1a35e5a71021fb79359bcfb472e66d4218bd8e09bb677e0d8be25957c3113a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07d7313454273e0a7f961fd172a7998afeefab6047ebaca57d79753d6055ca94", "generation_seconds_batch": 22.222612151992507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1970350742340088, "target_z": 1.648437420556917, "ce": 5.6254801750183105, "genuineness": 4.507347106933594, "identity": 0.6759917140007019, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5c1a35e5a71021fb79359bcfb472e66d4218bd8e09bb677e0d8be25957c3113a.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": "5cde75f538063d58a2b3235cac31f51a1b10c9a8897ac200d2d569519e89d2f2", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e21705ceefae11d2ae999112c5d195bae153f95aa042e82b9c21c40b2135b02f", "generation_seconds_batch": 21.846888614003547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30681470036506653, "target_z": -0.020507839017402694, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.3524791896343231, "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/emo_Confusion/5cde75f538063d58a2b3235cac31f51a1b10c9a8897ac200d2d569519e89d2f2.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "5d8537fa66495a8466a34de4f5afab97822fb1d661a3d96fe6d2f5c685e52cc3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f3c72535d1d92a5888ae8e3f69f54f0940845b906498a066959014df9b7b4980", "generation_seconds_batch": 20.67374404403381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.492907166481018, "target_z": 2.203125158886166, "ce": 5.808161735534668, "genuineness": 4.281409740447998, "identity": 0.3074489235877991, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5d8537fa66495a8466a34de4f5afab97822fb1d661a3d96fe6d2f5c685e52cc3.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "95b3637c693cd28d5e0b24ac0a6aa40da97a5452544ca7ededc1a71742f91efe", "generation_seconds_batch": 15.385275140986778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06355378031730652, "target_z": -0.47656250960301005, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.8274375200271606, "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/emo_Confusion/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "5ea2df131cf85942411c1c6bc77f43e755e0bcee784a704f5ba10574b8329bb2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ece1742e88d87db26f0747e518ec0e61f5eb6a01d194f384520a478daee603c", "generation_seconds_batch": 23.289554213988595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06876279413700104, "target_z": -0.46679688394825936, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.2882171869277954, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5ea2df131cf85942411c1c6bc77f43e755e0bcee784a704f5ba10574b8329bb2.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "5f34ee68156984d8cbfc54e413e9b966c0bf652a92c39750f714744dc46ec127", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "259277d0d431f3c3982cd81768cd09db311624045f06b81c23ca6b95319ebf21", "generation_seconds_batch": 22.114413153962232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6573812961578369, "target_z": 0.6367187032944512, "ce": 5.847874641418457, "genuineness": 4.243753433227539, "identity": 0.7595152854919434, "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/emo_Confusion/5f34ee68156984d8cbfc54e413e9b966c0bf652a92c39750f714744dc46ec127.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "5f413c8922ea3ddfdba2399315fbb86b9528521d7cdff4f3bf68902a081dae49", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "222d1c16dc96b82ef8d18a28dc058c18bd6927254e4da5b6ea6eb44aa2430686", "generation_seconds_batch": 20.842595033929683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43756094574928284, "target_z": 0.22460936212323654, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.36020177602767944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5f413c8922ea3ddfdba2399315fbb86b9528521d7cdff4f3bf68902a081dae49.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "5f4a7d3c133d1a4a7845e156a7b36fde73c2b2cb10640fe58de63d6e01f7d4c4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "712becf7f1e5609bf9f2cded75306a827d3aefefedccfce28fad370651a2d151", "generation_seconds_batch": 16.088130385032855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1302291601896286, "target_z": -0.3515624956349954, "ce": 6.1020402908325195, "genuineness": 4.846253395080566, "identity": 0.748288631439209, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5f4a7d3c133d1a4a7845e156a7b36fde73c2b2cb10640fe58de63d6e01f7d4c4.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "5fa181856f5ed6428d1cdfd71837555d28a66237721c52c691336824bb083167", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7439015d64231795bfbb2a26a6afd6ae80e726e48169dc6cdbfb8c8c4f07f0f1", "generation_seconds_batch": 21.17464775289409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7240567207336426, "target_z": 0.7617188010705533, "ce": 5.514282703399658, "genuineness": 4.6705241203308105, "identity": 0.7043874263763428, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5fa181856f5ed6428d1cdfd71837555d28a66237721c52c691336824bb083167.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "5feb4e20dd4cb43353f9dda35e57f6ef066da0256ec411ecd437f07cf22107aa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "08279331454378d018cb03515afc00b247157c136a5793519e07c54012972d53", "generation_seconds_batch": 16.38850259105675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0429787635803223, "target_z": 3.2343747363537245, "ce": 5.434855937957764, "genuineness": 4.369274139404297, "identity": 0.6734940409660339, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/5feb4e20dd4cb43353f9dda35e57f6ef066da0256ec411ecd437f07cf22107aa.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": "602dd3f514d7f29e16c3b922d005d489c8a50b2794a35191a0ec7a4ca5efa125", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "577dbf53208fec4ae06cb9f2018b7c1e6dee43f399cc888c5cb96106f46a6873", "generation_seconds_batch": 17.23186749301385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6512610912322998, "target_z": 2.5, "ce": 5.522225379943848, "genuineness": 4.337893962860107, "identity": 0.5497580766677856, "hypothesis": "Der Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du", "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/emo_Confusion/602dd3f514d7f29e16c3b922d005d489c8a50b2794a35191a0ec7a4ca5efa125.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "60791e6667a39fee92060699454177cdcd6ccac312a11bf4d7e1cbb6df77c9ff", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "146fc48370afaa1acd0d9c4b25f0678d7576f018fbbbe6ad87500f332d7396dc", "generation_seconds_batch": 15.126584465033375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42297571897506714, "target_z": 0.197265632638758, "ce": 5.7247633934021, "genuineness": 4.457139015197754, "identity": 0.688005805015564, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschwellen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/60791e6667a39fee92060699454177cdcd6ccac312a11bf4d7e1cbb6df77c9ff.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "60822df936e38b5e5e00fb0ecd1940500e7bcbccac590f3838bb8a27fec16f9e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "af2085b212c358265a35f090d0942b1af2ae234a9dbed0af00415875204076ce", "generation_seconds_batch": 21.51424367609434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30030345916748047, "target_z": -0.03271482219778999, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.4990958273410797, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/60822df936e38b5e5e00fb0ecd1940500e7bcbccac590f3838bb8a27fec16f9e.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": "60b4752dec458b1b0124ce18e2f6472f5396f6672166639447b2f30a533b0561", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c78d6be5d58bc010b7990e6bcc8515a0db921f14df3852d06b87c76827bcda49", "generation_seconds_batch": 15.908283584052697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 5.8875885009765625, "genuineness": 4.293961524963379, "identity": 0.7198479771614075, "hypothesis": "Dich schrieb die Nummer auf den Umschlag, lies 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/emo_Confusion/60b4752dec458b1b0124ce18e2f6472f5396f6672166639447b2f30a533b0561.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": "60ce25a812d282fba81720959f84f5ff61efd89e63c162c7f46c9c8a0ca22683", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3998c6f488cf3e96052f455c9a593191b33523f49d5a984dd44d206acdee88e9", "generation_seconds_batch": 20.30674255208578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1845334768295288, "target_z": 1.6249999860319855, "ce": 5.403085231781006, "genuineness": 4.256305694580078, "identity": 0.5165306925773621, "hypothesis": "Mes clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre, prends-les quand tu partir.", "wer": 0.08333333333333333, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/60ce25a812d282fba81720959f84f5ff61efd89e63c162c7f46c9c8a0ca22683.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "6169bbfd6f142601de38f247cf89d5eadc9d73166e98fba8bd3b404df637f201", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05499ccbfab6041a7a749829954238334331d849a926724cbeff69b70845fe58", "generation_seconds_batch": 21.746152767911553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7219730615615845, "target_z": 0.757812450238948, "ce": 5.990842819213867, "genuineness": 4.2249250411987305, "identity": 0.43662014603614807, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6169bbfd6f142601de38f247cf89d5eadc9d73166e98fba8bd3b404df637f201.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "62ae5e55de6ac2873ea565408b5ba75bcda799d15454e85930e0761f80ce9119", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ec193dc1e0d34947c9a42455308b719952521ec0689fe662894e5bdca9e7446a", "generation_seconds_batch": 23.383483270998113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19690454006195068, "target_z": -0.22656248166698084, "ce": 5.824047088623047, "genuineness": 4.758388519287109, "identity": 0.6627160310745239, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/62ae5e55de6ac2873ea565408b5ba75bcda799d15454e85930e0761f80ce9119.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6385d1033b2c0389381b3354053ae3f5542c7f8e4c159d57d7c563d910925a28", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "83f161ee35c3c668b8f317ff9d81ba4dd340d08b633f922fc5f0877b4750f1f4", "generation_seconds_batch": 21.756520421011373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04480132460594177, "target_z": -0.5117187731345242, "ce": 6.372091293334961, "genuineness": 4.557555198669434, "identity": 0.506793737411499, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6385d1033b2c0389381b3354053ae3f5542c7f8e4c159d57d7c563d910925a28.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "640da1a6f532799264d0d24dede81a005258b9c57fc09e33ab2aabcb295579b6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "451e562906267a7bcfad9e19dc4b3aa44c8d231cb89f3a28bfc39fc25748160a", "generation_seconds_batch": 20.762236740090884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 5.967014789581299, "genuineness": 4.231201171875, "identity": 0.5704396367073059, "hypothesis": "The door is still open? I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/640da1a6f532799264d0d24dede81a005258b9c57fc09e33ab2aabcb295579b6.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6443bb56c81c9171f8387c69e663f8a48ddb6d536305ac56160c2ed580142201", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f855e8c5913eb0c461020dce32988cef5c997a0e7fad8e72b8fbb622c6e21818", "generation_seconds_batch": 22.573375951964408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9094976186752319, "target_z": 1.1093750855540894, "ce": 6.0146708488464355, "genuineness": 4.2374773025512695, "identity": 0.4196857511997223, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6443bb56c81c9171f8387c69e663f8a48ddb6d536305ac56160c2ed580142201.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": "6469c16729c41f3219433ac9932e7ff551d1a7aaf3b80ca3f9ebea23b310383e", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9db299a5e50535e84ec2fd94f83cfe3c658fb20cbedab0d59dd470740ce6478f", "generation_seconds_batch": 16.27003201493062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.442900538444519, "target_z": 2.1093749738099725, "ce": 5.295859336853027, "genuineness": 4.344170093536377, "identity": 0.5479629635810852, "hypothesis": "Die Straße führt hinter der Brücke nach links. Wie weiter bis zu dem kleinen Holztor. 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/emo_Confusion/6469c16729c41f3219433ac9932e7ff551d1a7aaf3b80ca3f9ebea23b310383e.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "6489e1f1d54ebf00cc721877164bd2358f05695bd0d870e3a80c711c68e02ab5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ea7176785f9e79a1d76fcf1621982233da9bc77b06948292f07f3bc0ec1cf2b9", "generation_seconds_batch": 14.904567815014161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1593996286392212, "target_z": -0.2968750087300091, "ce": 5.982900142669678, "genuineness": 4.457139015197754, "identity": 0.6191796064376831, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6489e1f1d54ebf00cc721877164bd2358f05695bd0d870e3a80c711c68e02ab5.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "64e139a5df8bc00824353aa461764d655e02a09994b82dccec9b7a8a6f0921dd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a8b29818794f6f5b7d4ffa2c837d61537ad90f3d549beea69beb77ff963c4d49", "generation_seconds_batch": 16.033721885061823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6512610912322998, "target_z": 2.5, "ce": 5.422942161560059, "genuineness": 4.419482707977295, "identity": 0.611677885055542, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/64e139a5df8bc00824353aa461764d655e02a09994b82dccec9b7a8a6f0921dd.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "658b23b0e4fcd1dee1a360c08b0185f310f779423bfc0a70d46940860721f03d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3c410670522ebb35d95bb17edd240610544e7aad5e8d90001276f85fbd9c973b", "generation_seconds_batch": 19.504626156995073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.059647798538208, "target_z": 3.2656250960301003, "ce": 5.967014789581299, "genuineness": 4.268857479095459, "identity": 0.5078049898147583, "hypothesis": "De de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/658b23b0e4fcd1dee1a360c08b0185f310f779423bfc0a70d46940860721f03d.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "65f854603f9856a1f8f923fb99871685d24592806399fa31c2a4585a74a8a7fb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3bd3f21f8c1983c01bba13c692492a48ca9f062d3346dd993a15bce775ccae7d", "generation_seconds_batch": 20.898207806050777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.280379295349121, "target_z": 1.804687431032928, "ce": 5.458683967590332, "genuineness": 4.432034492492676, "identity": 0.6995426416397095, "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/emo_Confusion/65f854603f9856a1f8f923fb99871685d24592806399fa31c2a4585a74a8a7fb.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "66bd5227422fe371852a07864bd61f0e4bedffde5edb8a7dd9029a4c22a5e343", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dcefe6938e1b6a3066432379f48f168bc36b101050de4544df9f9d0f4aad0ffa", "generation_seconds_batch": 15.600008339970373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8636583089828491, "target_z": 1.0234376021411067, "ce": 5.824047088623047, "genuineness": 4.344170093536377, "identity": 0.591892659664154, "hypothesis": "Die Tür ist noch offen. Ich ich hab 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/emo_Confusion/66bd5227422fe371852a07864bd61f0e4bedffde5edb8a7dd9029a4c22a5e343.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": "670a130977a6a1c100ad96bc6c294901c5a556d03ffc14b33482211af661933b", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09c4046fdca176c090dab14ab47f36ed48a8c8fd37a42d59ea5bdcafe9b61755", "generation_seconds_batch": 21.623937588068657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7303075194358826, "target_z": 0.7734375183330191, "ce": 5.649308204650879, "genuineness": 4.532451152801514, "identity": 0.5472699403762817, "hypothesis": "La hôte tourne à gauche après le le pont continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/670a130977a6a1c100ad96bc6c294901c5a556d03ffc14b33482211af661933b.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "67271e6a012a7852e3fa33598e6dc1380501a7967396f07db0e0f3899e8ee8c4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a4845c86d48bd572cd5ac3fe42b5ccdc4509f8596f68d9d7e92b1e888d2d1883", "generation_seconds_batch": 23.630956389941275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.234670639038086, "target_z": 3.5937500733320764, "ce": 5.6254801750183105, "genuineness": 4.042920112609863, "identity": 0.5611446499824524, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/67271e6a012a7852e3fa33598e6dc1380501a7967396f07db0e0f3899e8ee8c4.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "677f88733208d109ff36ca6396a66a405f9ed146937ed4e9a1b942679b23bc67", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "846d777c36c25edc093c73fcd402508596eb71efa958f20145532a83f21d7fcd", "generation_seconds_batch": 14.839654676034115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0698046088218689, "target_z": -0.46484373646848587, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.599890947341919, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/677f88733208d109ff36ca6396a66a405f9ed146937ed4e9a1b942679b23bc67.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "67c841973174b1815d5effc53d85d7f62a904cc95e7a0c5d3946ec3693fb8c04", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "360433d6b9b68ec3b2f2d58e73b14ba479a0f4dadab768ce75ce0e559d4c968d", "generation_seconds_batch": 19.628857400035486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0513134002685547, "target_z": 3.2500001396801457, "ce": 5.561938762664795, "genuineness": 4.3504462242126465, "identity": 0.572267472743988, "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/emo_Confusion/67c841973174b1815d5effc53d85d7f62a904cc95e7a0c5d3946ec3693fb8c04.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "681581ea25bf4eabeaeecae67676847493035539bf48ee69bfe4f9c6b634711b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "241a0b58acdcb5594e9a8d3d46f4c84ef24266da181bbebae4578d9053ac0a7c", "generation_seconds_batch": 21.928588861017488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20628075301647186, "target_z": -0.20898437783725296, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.34829801321029663, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/681581ea25bf4eabeaeecae67676847493035539bf48ee69bfe4f9c6b634711b.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": "6826d43249bbb78c53ae2f86bf5e2d958ab95408f7cf900ad5b67937957174aa", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Your message arrived after I had left.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70dff0ea5df654efcd6fcbedddb78e2342e0620e1c74a96946820aa98110ae75", "generation_seconds_batch": 16.22802845109254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7782304286956787, "target_z": 0.8632812408334904, "ce": 5.831989288330078, "genuineness": 4.231201171875, "identity": 0.2171601504087448, "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/emo_Confusion/6826d43249bbb78c53ae2f86bf5e2d958ab95408f7cf900ad5b67937957174aa.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "684eb51cee7dc6ab5b100d952f7bb79b1d16d748b63bcdea44e73e5c517f366e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e23edabacbab95d76e0ede9a6a4ef1408d018040d8951ef33058728125568d86", "generation_seconds_batch": 21.218752069980837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 6.205295085906982, "genuineness": 4.218649387359619, "identity": 0.38314568996429443, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/684eb51cee7dc6ab5b100d952f7bb79b1d16d748b63bcdea44e73e5c517f366e.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "6a27269a255cba89b8ed28b46ddb942f200c0ce21410927ef29560e24f784804", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21e0160d44a184185522465f7533581c5b0b1d7c8ec657e481e285f9439b8afa", "generation_seconds_batch": 20.075562836020254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06563737988471985, "target_z": -0.47265627051552145, "ce": 6.348263263702393, "genuineness": 4.15588903427124, "identity": 0.18864230811595917, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6a27269a255cba89b8ed28b46ddb942f200c0ce21410927ef29560e24f784804.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ac3d02232904fd9dd77831c6dd9a3bcc939ee60ea8efb5c73b95e77f8614c2c8", "generation_seconds_batch": 16.2280199440429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07188820838928223, "target_z": -0.46093749738099726, "ce": 6.41180419921875, "genuineness": 3.9927117824554443, "identity": 0.27108803391456604, "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/emo_Confusion/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "6ab433bae79d2f38a919f7c47f92adf798fc2f411cf7248f130dfec5c9e2cd0c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "61cacffdc7016cc678f66b0cc5a57d3d3d1208064964597ccb5219f9a2d056b1", "generation_seconds_batch": 15.0956993320724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3208790421485901, "target_z": 0.005859358631232906, "ce": 5.982900142669678, "genuineness": 4.68307638168335, "identity": 0.6192072629928589, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6ab433bae79d2f38a919f7c47f92adf798fc2f411cf7248f130dfec5c9e2cd0c.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "6aecbef5722f61c069224bdad2bea73bde09dcfcfc9f7581a8fb9679dd276df6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "322eb62abd449d4bcb34f3b54e17b5df0c31e2d6922458a5a6ec7ab0b24aedfc", "generation_seconds_batch": 20.419816776993684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09480786323547363, "target_z": -0.4179687556745059, "ce": 5.998785495758057, "genuineness": 4.833701133728027, "identity": 0.6478434801101685, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6aecbef5722f61c069224bdad2bea73bde09dcfcfc9f7581a8fb9679dd276df6.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "6aece54569cbed94676671cbcaa25cbf376be894f64c1f460b4f220f23dc34e2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f8cfc938a792bc081bb827c9699909e70b50bb2a7f0566d8cca7112f86d63e31", "generation_seconds_batch": 21.301815377897583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2303727865219116, "target_z": 1.7109374694449682, "ce": 5.847874641418457, "genuineness": 4.356721878051758, "identity": 0.4945344030857086, "hypothesis": "Ik heb het een klein doosje gevonden. Het stond naast het oude raam. We kunnen daar 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/emo_Confusion/6aece54569cbed94676671cbcaa25cbf376be894f64c1f460b4f220f23dc34e2.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "6b100a23965964428559c03136d84216214049af33f9105f3fc4a245ce1784cb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5309a7fff3b55b28168b1e003174f358c24bb2f463040aa82224414ddc9c5d5d", "generation_seconds_batch": 20.768599046976306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9929723739624023, "target_z": 3.140624998253998, "ce": 5.77639102935791, "genuineness": 4.21237325668335, "identity": 0.7372264862060547, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6b100a23965964428559c03136d84216214049af33f9105f3fc4a245ce1784cb.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "6b7ba74ec098bcee1d99d7200fdbeca705a46c705bc000fa79ff83ccc96872c2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "441784cada2f43ca4d20a27c2b80239085beb403bec669ce29d7247b6eca7a41", "generation_seconds_batch": 20.618317262036726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2970482110977173, "target_z": 1.8359375672210703, "ce": 5.990842819213867, "genuineness": 4.331617832183838, "identity": 0.7259138822555542, "hypothesis": "Ik heb een klein doosje 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/emo_Confusion/6b7ba74ec098bcee1d99d7200fdbeca705a46c705bc000fa79ff83ccc96872c2.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "21b6f95c0972910ea4345bb8a20ab3ed296cec5711bb92a21783dadbf1d751c2", "generation_seconds_batch": 15.091096078976989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.205295085906982, "genuineness": 4.557555198669434, "identity": 0.6220277547836304, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6ba802036bef4c46b063b6a7877be6e2cf978a1e780e3e976accd56726111026", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0c4f666e3058fd13e26b7d6a18066945db81fd6be2f0fce516895bab5de2080b", "generation_seconds_batch": 21.47277455602307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16044144332408905, "target_z": -0.2949218612502356, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.2654363811016083, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6ba802036bef4c46b063b6a7877be6e2cf978a1e780e3e976accd56726111026.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "6baad34139d733abab27ef3477f4fcbcf04dcee5d2d506659a07c333a962b44c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f17046e046543ac140e9aa74ed8c1ba213a16f1801cb3fa3d39f5af58f5a823c", "generation_seconds_batch": 16.738844127976336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0698046088218689, "target_z": -0.46484373646848587, "ce": 6.332377910614014, "genuineness": 4.7834930419921875, "identity": 0.6542339324951172, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6baad34139d733abab27ef3477f4fcbcf04dcee5d2d506659a07c333a962b44c.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "6bf17da0dbc6237872077223dddcf4b3366ee865aa044c3d8d3c97124f0e6d52", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a958f982e3e459f8227f1ab2f84bd23089af66ae538ad1e6c8380b9176419c9c", "generation_seconds_batch": 22.168654451030307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.443031072616577, "target_z": 3.9843748760338706, "ce": 5.522225379943848, "genuineness": 4.042920112609863, "identity": 0.09519977867603302, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6bf17da0dbc6237872077223dddcf4b3366ee865aa044c3d8d3c97124f0e6d52.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "6cf53503db661abc978194c24d22111b3d70e23a4e6b9d205f1ebe490a7614d0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "271d0267289c1417274f2e53295779171cb074eb377cee979d027cb5029c9abb", "generation_seconds_batch": 29.028887607040815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2428743839263916, "target_z": 1.7343749039698997, "ce": 5.871703147888184, "genuineness": 4.20609712600708, "identity": 0.47801417112350464, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6cf53503db661abc978194c24d22111b3d70e23a4e6b9d205f1ebe490a7614d0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "6d0518611a08d41762ab5ed401f727e3119ee42e9dcf383c58c2762b48b9d03d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53cb1c914a215e9c0014833146c4fc5372052f56d33a121ff7f1c813f0f41b64", "generation_seconds_batch": 16.356361845042557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0761860609054565, "target_z": 1.4218751065061113, "ce": 5.839932441711426, "genuineness": 4.432034492492676, "identity": 0.6646347641944885, "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/emo_Confusion/6d0518611a08d41762ab5ed401f727e3119ee42e9dcf383c58c2762b48b9d03d.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": "6d41a186d1df44f20aad65687145277c215513fc7469ec2bc8ffa70c1f8b563f", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d9b9adcead194decd594219839d7916075c4a4a04f198168861002325d74808a", "generation_seconds_batch": 14.405039991019294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8678255081176758, "target_z": 1.0312500803160838, "ce": 5.605623245239258, "genuineness": 4.645420074462891, "identity": 0.7806365489959717, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue Elle était ici près de la grande bibliothèque Nous la remettrons après le déjeuner", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6d41a186d1df44f20aad65687145277c215513fc7469ec2bc8ffa70c1f8b563f.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "6ddc056b33d1a68a415bf25171b51225ee15a3ac1d47413cb07351cff88fad69", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f856454277b7eb04ee9f556ef9fa1734d73af994603a3f44b6e531034b0287b8", "generation_seconds_batch": 22.14455138600897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4094322919845581, "target_z": 0.1718750226980237, "ce": 5.117149353027344, "genuineness": 4.356721878051758, "identity": 0.6339424252510071, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6ddc056b33d1a68a415bf25171b51225ee15a3ac1d47413cb07351cff88fad69.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6defb30d360a168e96883cc20b4005249b01ef5470fea0c42a59b0d517ff88cc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "58c0aff518aabce4815e6037174bfe7bd1cc89ae78c07b6bafcaed191a8234b7", "generation_seconds_batch": 21.71078587102238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1136908531188965, "target_z": 1.492187410080906, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.5192726850509644, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6defb30d360a168e96883cc20b4005249b01ef5470fea0c42a59b0d517ff88cc.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "6e33382feb2003423ca6b2212dace92b89598460047fda3560ddaeddb386bf8d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "fcb14697944bf4d7bb19fdb68b4056201ac39d4834011f3548576e9fcaa048b5", "generation_seconds_batch": 19.924826035043225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.667930006980896, "target_z": 2.531250136188142, "ce": 5.617537498474121, "genuineness": 4.645420074462891, "identity": 0.7146301865577698, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6e33382feb2003423ca6b2212dace92b89598460047fda3560ddaeddb386bf8d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "6e8b6e6ae9684ecb2225389f23f92e4dd1a2de3cac9bf6df0ae06c6d4398c0f0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e45d611606b0dd85dd9700b25ba39b573e37d01c436278e7740a80918a252008", "generation_seconds_batch": 22.009176153922454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6240436434745789, "target_z": 0.5742187661505168, "ce": 5.506340026855469, "genuineness": 4.369274139404297, "identity": 0.6783026456832886, "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/emo_Confusion/6e8b6e6ae9684ecb2225389f23f92e4dd1a2de3cac9bf6df0ae06c6d4398c0f0.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "6ede84d5ee5a8e881a121bb2e1865db6d969e7aae88c5887832c049a487c241b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4ab8f88cd3527c6603ef381594079a86f282f87e2d2724ab1934f1fcb870fdce", "generation_seconds_batch": 14.691540786065161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.6620147824287415, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6ede84d5ee5a8e881a121bb2e1865db6d969e7aae88c5887832c049a487c241b.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "6eeddf295027320c5d82028cb0d91578d74819bc8b82df0f26090bfd6748ac70", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3bcf4c8bf9707be93d51633392104876ae7ec0534d5f38906eb624f02c8b311e", "generation_seconds_batch": 21.691767125972547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1887006759643555, "target_z": 1.6328124642069626, "ce": 5.974957466125488, "genuineness": 4.344170093536377, "identity": 0.5116350054740906, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6eeddf295027320c5d82028cb0d91578d74819bc8b82df0f26090bfd6748ac70.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": "6ef5d7ea523ae72295e2064f7bd6adf3fd22d4bf96f068f07e546953ddf62a63", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1ba209219b3ce49172f7363e960abe174732c1e29d232a25e1e82f1cf740171", "generation_seconds_batch": 21.921301924041472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.318014621734619, "target_z": 3.7499996368316206, "ce": 5.704906940460205, "genuineness": 4.331617832183838, "identity": 0.6266042590141296, "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/emo_Confusion/6ef5d7ea523ae72295e2064f7bd6adf3fd22d4bf96f068f07e546953ddf62a63.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "6f00f713804a240b1932a8f19fb565e0efb9143858baf3e04c7660a77fac319d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "81173deaac5bd693a61bb4c52944eff01aed6b2ecf99f8ecef797b3621d6d2a3", "generation_seconds_batch": 19.58181109803263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13543817400932312, "target_z": -0.3417968699802448, "ce": 6.141753673553467, "genuineness": 4.796044826507568, "identity": 0.6362650394439697, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6f00f713804a240b1932a8f19fb565e0efb9143858baf3e04c7660a77fac319d.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "6f0533ef5412db94dd131c3e3213e4f5faa46449fa649c28518be2818c2289eb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b258e502ac083e31cc72697640483e1406a8e9e9baeadbc2c72471fa34763343", "generation_seconds_batch": 16.48558146005962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.584455132484436, "target_z": 0.5, "ce": 5.967014789581299, "genuineness": 4.369274139404297, "identity": 0.38209688663482666, "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/emo_Confusion/6f0533ef5412db94dd131c3e3213e4f5faa46449fa649c28518be2818c2289eb.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "6f3f5ede2c4bfed0369655ef78048a63f0b50cd56ba051de177273fcc9a8d57d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d34c6c774d9d081a2fdbef76db8558f65d993e4eaae32312309af8a8ebef924", "generation_seconds_batch": 23.6033853800036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5595824718475342, "target_z": 2.3281250331740346, "ce": 4.905179023742676, "genuineness": 4.262581825256348, "identity": 0.6353959441184998, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6f3f5ede2c4bfed0369655ef78048a63f0b50cd56ba051de177273fcc9a8d57d.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "6f6ebba792f8dfd1c28e9a7cdcf8e7a6c48da820f0d151fb045835176d3ff614", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5a89b279f5f2d044c9ffd2a7874cdb8da44fb4adc44791cb9b5a0d5c391d8035", "generation_seconds_batch": 19.616920884000137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11043490469455719, "target_z": -0.3886718787102539, "ce": 5.720792293548584, "genuineness": 4.934117794036865, "identity": 0.7450689673423767, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6f6ebba792f8dfd1c28e9a7cdcf8e7a6c48da820f0d151fb045835176d3ff614.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6f7f243c4ad400cdc59358495cbf8e31e6b6a650dcb8e60b2d751642c0d04515", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ad0d1529ad8c54857a255482b89b2b6a78d67224b923a9211db0f3e08fcce654", "generation_seconds_batch": 21.756520421011373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16669225692749023, "target_z": -0.28320311605174064, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.34159234166145325, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6f7f243c4ad400cdc59358495cbf8e31e6b6a650dcb8e60b2d751642c0d04515.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "6fc6c3e1eb057c362e6b1ead43cd1461fdd262b2fc5249adcbf0e763c5e1a09e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "448002e53bc40dd4ecf2da809e5e4ab48be76f8a9777f35d4c803c7407eaaebd", "generation_seconds_batch": 20.374852536944672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3247857987880707, "target_z": 0.013183570888288624, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.6494166851043701, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6fc6c3e1eb057c362e6b1ead43cd1461fdd262b2fc5249adcbf0e763c5e1a09e.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "6fdb3c72f17412db28f846a06e5596ce8ab6922aa974be0827aedda046049458", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2664a2d73ffa418186b73668c38e42229993971b2f33ca858fc2a4b421dba7d", "generation_seconds_batch": 18.816177662927657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5115289688110352, "target_z": 0.36328129670554876, "ce": 6.117925643920898, "genuineness": 4.557555198669434, "identity": 0.532325267791748, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/6fdb3c72f17412db28f846a06e5596ce8ab6922aa974be0827aedda046049458.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "706cac8a8a403475d9076efc623e2c5be36b7524574bae8200fec1ebfdabc002", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "48df2752b7c8854af3b041c54a9f78673d952a6ba59e0e892c0dddba197455c6", "generation_seconds_batch": 27.377952252049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4679038524627686, "target_z": 2.1562500663480693, "ce": 5.486483573913574, "genuineness": 4.532451152801514, "identity": 0.34919974207878113, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/706cac8a8a403475d9076efc623e2c5be36b7524574bae8200fec1ebfdabc002.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "70ced3eaa0ba48eb4deec53b8ad7db2caaae6828b78afc5db4c0fe1f8a8376bd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e4bfb70fc8079bf18d90f036a6e6b24de4f5c1c883a28f11ee570053e5114b4", "generation_seconds_batch": 16.289375386899337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9845073223114014, "target_z": 1.2499999161919124, "ce": 5.272031307220459, "genuineness": 4.532451152801514, "identity": 0.6999265551567078, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/70ced3eaa0ba48eb4deec53b8ad7db2caaae6828b78afc5db4c0fe1f8a8376bd.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "719188ee27cd9991eb04c1e923de62ae80fd9eb476616458f89348f8698f6dd1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "41f21f75b505df6c51e9e6f831109b789a11e3ca2bb0c4ab76decc9159917d60", "generation_seconds_batch": 27.409488159930333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7969828844070435, "target_z": 0.8984375043650046, "ce": 5.530168056488037, "genuineness": 4.532451152801514, "identity": 0.6554187536239624, "hypothesis": "La portée est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/719188ee27cd9991eb04c1e923de62ae80fd9eb476616458f89348f8698f6dd1.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "71a25bdb765f743080f27bd7d3358d252c36ee12858d533d0b62beeedec5b31d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0c70969aeaece957caeb5815bf800b86532a6eacd3320b5c4703a52df1528de1", "generation_seconds_batch": 20.342482059961185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2763426303863525, "target_z": 3.6718748550818487, "ce": 5.546053409576416, "genuineness": 4.268857479095459, "identity": 0.3858296871185303, "hypothesis": "De deur 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/emo_Confusion/71a25bdb765f743080f27bd7d3358d252c36ee12858d533d0b62beeedec5b31d.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "71deae60b4238252b8861f4daff0c9f26cda7ba6b16627692499132da2f3a1fb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "fc5356b6b936e048ed7e413ae0129108440c52e03addf63adc8661b277844be1", "generation_seconds_batch": 17.154648658935912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10835130512714386, "target_z": -0.3925781177977425, "ce": 6.24500846862793, "genuineness": 4.7081804275512695, "identity": 0.7988153696060181, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/71deae60b4238252b8861f4daff0c9f26cda7ba6b16627692499132da2f3a1fb.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "720085128aad2308d539b5b82d8f8e2276687f06353338940631d159b56445e4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "243ea5ab83d2ecbba46111aac71776f95ac9e95ebb23953077d82ab4766d2dba", "generation_seconds_batch": 20.238970909966156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6240436434745789, "target_z": 0.5742187661505168, "ce": 6.133810997009277, "genuineness": 4.3504462242126465, "identity": 0.23226769268512726, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/720085128aad2308d539b5b82d8f8e2276687f06353338940631d159b56445e4.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": "72147b61925606f4e446a95ee34f1cb1485194e3c7a24b2ba5a99ae31ef13c8b", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "229c5d553ab3a2d24a84e2097f74529065406ec7b408c3ba90b404502b5b7603", "generation_seconds_batch": 15.57283340091817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9262970685958862, "target_z": 3.0156251239661294, "ce": 5.514282703399658, "genuineness": 4.3881025314331055, "identity": 0.6387437582015991, "hypothesis": "Deine Nachricht kam sich schon weg vor. Ich ich lass sie zweimal, dann leg dich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.2916666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/72147b61925606f4e446a95ee34f1cb1485194e3c7a24b2ba5a99ae31ef13c8b.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "7228f06c09211742253ff03d2583682afcfbe14ca757698160433759dcd8bb81", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9bbe922d1eb4c671f1673bce2f3b72d4830ba3b091d3a5b67f9455d06f26e934", "generation_seconds_batch": 15.683584069018252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.617923378944397, "target_z": 2.437499951111949, "ce": 5.657250881195068, "genuineness": 4.344170093536377, "identity": 0.7118936777114868, "hypothesis": "Ja, ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "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/emo_Confusion/7228f06c09211742253ff03d2583682afcfbe14ca757698160433759dcd8bb81.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "723482d695d09a2943ed2de7dacc7f359f4465523c48c97ece41533f8ccf7dc5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1e176203743d669fe47408ab6e8812acb5b05680440d91482cd28b473a1d10f9", "generation_seconds_batch": 19.625613541924395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8678255081176758, "target_z": 1.0312500803160838, "ce": 5.8875885009765625, "genuineness": 4.519899368286133, "identity": 0.6463335156440735, "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/emo_Confusion/723482d695d09a2943ed2de7dacc7f359f4465523c48c97ece41533f8ccf7dc5.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "7246b4871e1f069446b2eefe5f6d4d241020498dce21ec319678e8682e9d7bd5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ec4d48993fadf3504ac40e15e5ab47dacf5d20e5f91025057e381f589f087b1", "generation_seconds_batch": 20.761759391985834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.47714945673942566, "target_z": 0.29882812827375344, "ce": 5.935244083404541, "genuineness": 4.65797233581543, "identity": 0.5147385597229004, "hypothesis": "He encontrado la cajita. Estaba. Esté ahí junto 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/emo_Confusion/7246b4871e1f069446b2eefe5f6d4d241020498dce21ec319678e8682e9d7bd5.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "7255fe060530ddf13459bf6fad6b2c2d77dd7f4621f849760d13490cb1c1be4e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f21b38f41d1b2561b07c2c38c2be72fd5a77fd8f28d160a366ed01f4fe9207f5", "generation_seconds_batch": 19.162064046016894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.734605312347412, "target_z": 2.6562500104760107, "ce": 6.054384231567383, "genuineness": 4.193545341491699, "identity": 0.5557152032852173, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7255fe060530ddf13459bf6fad6b2c2d77dd7f4621f849760d13490cb1c1be4e.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "72ade2508e39b97fbdc4cf359a7fc0b01626d7b13fac1d3b6f296174b7bfb4dd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "11678133adb94225be8ffee04420a70e67730a3b56ef26a860a4d7d9c69d851b", "generation_seconds_batch": 16.290092927054502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.070269584655762, "genuineness": 4.532451152801514, "identity": 0.7524391412734985, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/72ade2508e39b97fbdc4cf359a7fc0b01626d7b13fac1d3b6f296174b7bfb4dd.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "72fd8954e448e53a5d79a6a8fd8aa50fca2267c422cb43e3fa45012d5069cc34", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eef0b2b9f4518139ea27eaca0374425b8ec83d01778058f002461dfc6aeed363", "generation_seconds_batch": 20.047504008980468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.2479628324508667, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/72fd8954e448e53a5d79a6a8fd8aa50fca2267c422cb43e3fa45012d5069cc34.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "733e54ab00c7c422d62aaa87a17e337a25c911434cd05219d1e0786f4e4dad43", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0e8d08bccf9bc610fcf3f60f51a750cd2eb81114980e033c92eefc3407f6b990", "generation_seconds_batch": 20.624740824918263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.213703989982605, "target_z": 1.6796875567450593, "ce": 5.911416053771973, "genuineness": 4.482243061065674, "identity": 0.6549729108810425, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken hoor.", "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/emo_Confusion/733e54ab00c7c422d62aaa87a17e337a25c911434cd05219d1e0786f4e4dad43.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "737525b81b0a53e8f312244ed0fbb8c171c615d5ac095a733c6bd238c4830969", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "42f56ef8b634d7a55d32b0e846dafc5c24f53c90be73826358135944cf540026", "generation_seconds_batch": 20.732215017080307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3013458251953125, "target_z": 3.718749724131712, "ce": 5.77639102935791, "genuineness": 3.948779582977295, "identity": 0.6180892586708069, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/737525b81b0a53e8f312244ed0fbb8c171c615d5ac095a733c6bd238c4830969.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "74222034d3b6d53bd7382ba9b16b67c4f649621409cbcbd07be5bdcc6cadeecb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c4f9d36695e1a644b16ea19bebb5c3b8f7da25b010aabb3f0688cd47d300ea08", "generation_seconds_batch": 20.364825114025734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4490208029747009, "target_z": 0.24609378884854058, "ce": 6.157639026641846, "genuineness": 4.7834930419921875, "identity": 0.7096633911132812, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/74222034d3b6d53bd7382ba9b16b67c4f649621409cbcbd07be5bdcc6cadeecb.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "742cecd2f9e49b460cf14d3020389e6173430155fbf13c7bd67e202fdf577b39", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0060d1e1177f30596a232737c8b311ccf441ea237a0f4b30ae59912dd5e9a830", "generation_seconds_batch": 14.954697004053742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6928026080131531, "target_z": 0.7031249912699908, "ce": 6.086154937744141, "genuineness": 4.406930446624756, "identity": 0.7209631204605103, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/742cecd2f9e49b460cf14d3020389e6173430155fbf13c7bd67e202fdf577b39.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "74dcc2864dade9c995e6f11a7ef39219faf4b2319d86802364043b293464c627", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8c26c253e5467341ee690b5c6b21db5f294d55b7c8c8f21c54d47968ee1147c", "generation_seconds_batch": 21.201141183031723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.671966552734375, "target_z": 0.6640624886509882, "ce": 5.903473377227783, "genuineness": 4.18726921081543, "identity": 0.41536396741867065, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/74dcc2864dade9c995e6f11a7ef39219faf4b2319d86802364043b293464c627.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "758c2c423b89c1fd2c67e973d7b6628e677190930c54cd65547604842778e47f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad9bff9bb68de74454106aa5780050444d7583b79743065cd88a007e20ca3129", "generation_seconds_batch": 21.811032108031213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.126192569732666, "target_z": 1.515625068094071, "ce": 5.498397350311279, "genuineness": 4.482243061065674, "identity": 0.5944725275039673, "hypothesis": "J'ai 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/emo_Confusion/758c2c423b89c1fd2c67e973d7b6628e677190930c54cd65547604842778e47f.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "76d68ca863d75548e5590d38f2f86b98e3e16cd0bc2137580564ac946b4d28e8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8e1cb89cd0a8fbb311bf90cf8af0266fb416b2803b5c80934312cfebeccf0ca3", "generation_seconds_batch": 19.892604740918614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.086154937744141, "genuineness": 4.331617832183838, "identity": 0.5623838901519775, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/76d68ca863d75548e5590d38f2f86b98e3e16cd0bc2137580564ac946b4d28e8.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "76e784cdbaa5271a20a016c25e407f5f2683761bd2f95eaf46605c103cf99dcc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "63087c1907dc178c2cb06be0954d1f5a9565d787f96983000890c34261a4de97", "generation_seconds_batch": 22.060853286995552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3422360122203827, "target_z": 0.04589844895813697, "ce": 6.078212261199951, "genuineness": 4.65797233581543, "identity": 0.4267652630805969, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/76e784cdbaa5271a20a016c25e407f5f2683761bd2f95eaf46605c103cf99dcc.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "76eeb129b341839829fc9e72a991be7cc376adc62df64d3c6e757d1c4a21ee50", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d237b84f603ea65cdc72057d87c3f57f74e89dc4d2b5e57fdefb6e03d5347d1d", "generation_seconds_batch": 21.92049885704182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12606194615364075, "target_z": -0.35937500174600184, "ce": 5.935244083404541, "genuineness": 4.758388519287109, "identity": 0.6547706723213196, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/76eeb129b341839829fc9e72a991be7cc376adc62df64d3c6e757d1c4a21ee50.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "29a35ea2b60f5ce684e4063100de5ad85fe1a95b222cf9a40bc33e802bc4196e", "generation_seconds_batch": 15.964063439052552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.078212261199951, "genuineness": 4.356721878051758, "identity": 0.651509702205658, "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/emo_Confusion/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "779dc1a81b0b3701b575617a4dd88df34f4718a7161682a9112da5e9b4f5c37d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bf9faa41b91c956a00a86d310205398d6c357f267aca8fce53c506d3566479df", "generation_seconds_batch": 20.880221002968028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7719796299934387, "target_z": 0.8515625235710246, "ce": 5.7684478759765625, "genuineness": 4.457139015197754, "identity": 0.6277347803115845, "hypothesis": "La porte est encore y verte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/779dc1a81b0b3701b575617a4dd88df34f4718a7161682a9112da5e9b4f5c37d.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "7836d76e1b95972d776761fa679fd1b0e0634bb94b2c73f7b4c9a5c86748a289", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afa117d0ac6504e36941848468b5d80878ed2cba8798bdf02117e8ec24797539", "generation_seconds_batch": 17.45860060909763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06147018074989319, "target_z": -0.48046874869049866, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.299615740776062, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7836d76e1b95972d776761fa679fd1b0e0634bb94b2c73f7b4c9a5c86748a289.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "7882157457a3124defde38b47f44e6f8d0cc67b8d2316cf9bd88ee045de9edb3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "542e82dcbbd26cb3fcbc7bc2bb78bec9f323186108d2d8744406d13a432792f8", "generation_seconds_batch": 21.69620088103693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.428705632686615, "target_z": 0.20800781806538082, "ce": 5.824047088623047, "genuineness": 4.545003414154053, "identity": 0.5745134949684143, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7882157457a3124defde38b47f44e6f8d0cc67b8d2316cf9bd88ee045de9edb3.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "788ec12c511ec9c14f534a8d64107644a134847f130549bc06ef5cae1af696df", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f37f68db9ae2856318ae7a117ebd3119ea7d596661ac28dbcf4a09d0b4ee810b", "generation_seconds_batch": 15.93824567401316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.238707184791565, "target_z": 1.7265624257949226, "ce": 6.0146708488464355, "genuineness": 4.570107460021973, "identity": 0.55415278673172, "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/emo_Confusion/788ec12c511ec9c14f534a8d64107644a134847f130549bc06ef5cae1af696df.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "790fbb3eef18982ca7b1eb5e0400081d77af9ed270b1b5e48256949a5542e1e2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c630ba3d49a56599b9c97b370999d8d907fa74284b85b1160c73e98456cee2b7", "generation_seconds_batch": 19.962014107964933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7929463386535645, "target_z": 2.765625151902159, "ce": 5.218418121337891, "genuineness": 4.055472373962402, "identity": 0.25871923565864563, "hypothesis": "Ik heb het een klein doosje gevonden, van het stond naast het oude raam. Kunnen we samen in kijken?", "wer": 0.2777777777777778, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/790fbb3eef18982ca7b1eb5e0400081d77af9ed270b1b5e48256949a5542e1e2.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "7921474f417f5a5ad61fdddd9ff7edb227b081193d8f071bf5974a2298b0845c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f010207fc9420a204a6909ab67434a40fa98bd2924aec2692296e0ede8140615", "generation_seconds_batch": 20.509842209983617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04480132460594177, "target_z": -0.5117187731345242, "ce": 6.157639026641846, "genuineness": 4.262581825256348, "identity": 0.1489294469356537, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7921474f417f5a5ad61fdddd9ff7edb227b081193d8f071bf5974a2298b0845c.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "793327736ac819bcd39857a0b212185e054825aa0ea4651ad09202bbf043eea1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5eaaf5a49c58703c340e409491a2f9fd27b59e4e8fd7800289d57e6ab014c93", "generation_seconds_batch": 15.852434939937666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1708594560623169, "target_z": -0.27539063787676343, "ce": 5.967014789581299, "genuineness": 3.973883628845215, "identity": 0.22703944146633148, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/793327736ac819bcd39857a0b212185e054825aa0ea4651ad09202bbf043eea1.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "7943590b58c69ebfd419c899897ac1b62d96f3688c831f37811110204a873021", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "eb08e2f4a726b100e8b576b54efc761b3faeac64efc28d874b0ce765cdd270a8", "generation_seconds_batch": 16.41328211594373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0178450345993042, "target_z": 1.3124999650799636, "ce": 5.998785495758057, "genuineness": 4.413206577301025, "identity": 0.3727075457572937, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7943590b58c69ebfd419c899897ac1b62d96f3688c831f37811110204a873021.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ea96d03f44f9ed0cbcf0c239d0133aff44d737df3f1dbe7d262c44840782d06b", "generation_seconds_batch": 14.94309678603895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.976172924041748, "target_z": 1.234374959841958, "ce": 6.38797664642334, "genuineness": 4.519899368286133, "identity": 0.6591747999191284, "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/emo_Confusion/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.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": "7958d45a766252f936334db5192989d6bc2e5639a4030296ae5f673d666f7dbc", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "badb3d62593375f412a31a55d1231c58127e987bd0acd71a724fe123c35002ca", "generation_seconds_batch": 15.57283340091817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1763296127319336, "target_z": 3.4843749319059287, "ce": 5.6413655281066895, "genuineness": 4.532451152801514, "identity": 0.6583086252212524, "hypothesis": "Und eine Nachricht kam, als ich schon weg war. Ich lasse sie zweimal, dann leg dich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.20833333333333334, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7958d45a766252f936334db5192989d6bc2e5639a4030296ae5f673d666f7dbc.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "79749fa5316368a028fae9878b12348e5ee8b74451885e5aec4677f7d3841169", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8c3e089941decf0babaeb38a442b7b7e80942221ba1be2df3e2dc8544e4b1935", "generation_seconds_batch": 21.735346120898612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3213999569416046, "target_z": 0.006835946339134231, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.5923367738723755, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/79749fa5316368a028fae9878b12348e5ee8b74451885e5aec4677f7d3841169.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "79f9c810541bc222d808a1caa011a555bad0fb7fcfde1c1adfc8c75d8f2a91e3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2322104c654c71e5946eb3f2bf6fc4992d0196edd8960a1947b785b762bd8bd7", "generation_seconds_batch": 21.911486500059254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6512610912322998, "target_z": 2.5, "ce": 5.450741291046143, "genuineness": 4.400654315948486, "identity": 0.7981635928153992, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/79f9c810541bc222d808a1caa011a555bad0fb7fcfde1c1adfc8c75d8f2a91e3.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "7a05fc6d925d9b1ab032ab8b2c1d81c3c20b6436b3caf9970a79afecaaabb9ea", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0e4b032bc3543dc43cc6cf373f440f0173a933511aaacd287174010ef4539d5e", "generation_seconds_batch": 14.893556658993475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.509445309638977, "target_z": 0.3593749458739435, "ce": 6.086154937744141, "genuineness": 4.356721878051758, "identity": 0.21674925088882446, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7a05fc6d925d9b1ab032ab8b2c1d81c3c20b6436b3caf9970a79afecaaabb9ea.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "7a2473fd420f49993ba50dd402a2bb22d1eabef0c2a90740973ef73007b451ba", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "640d72f431a4ec99aa6999cf5bcc51262bf536fa8229608f60215534eacbb9f4", "generation_seconds_batch": 16.356361845042557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43651914596557617, "target_z": 0.22265624257949226, "ce": 5.95112943649292, "genuineness": 4.595211505889893, "identity": 0.6970819234848022, "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/emo_Confusion/7a2473fd420f49993ba50dd402a2bb22d1eabef0c2a90740973ef73007b451ba.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "7a79ceea987b405af62a38ae710293a0dd831475f40ab560e8e8a45634e56910", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0c1606b904ba773337f069923b161efb447f3fcc7664df13f20b8ce410656c1f", "generation_seconds_batch": 21.047677561058663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.221180438995361, "genuineness": 4.300237655639648, "identity": 0.5667504668235779, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7a79ceea987b405af62a38ae710293a0dd831475f40ab560e8e8a45634e56910.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": "7a9cf47476d56db614062139219d9c6d56b847c7fb9409fa59415bc7513360b4", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca69d6a6a50138f8f3889aacf0133b651733221fb11a9c1ef39a448c96333767", "generation_seconds_batch": 14.641263417084701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8346184492111206, "target_z": 2.843750157140164, "ce": 5.7128496170043945, "genuineness": 4.432034492492676, "identity": 0.8201324939727783, "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/emo_Confusion/7a9cf47476d56db614062139219d9c6d56b847c7fb9409fa59415bc7513360b4.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "7b334fd3cfb94fdd5f62f762219e7160df611b5039ef0b2a8bd2d6896bfe44ed", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af869fb722a34ccb23ae301765d078f831dc6dd68a82ec89c5a4544351da7df6", "generation_seconds_batch": 15.852434939937666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07397180795669556, "target_z": -0.45703125829350866, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.2751125693321228, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7b334fd3cfb94fdd5f62f762219e7160df611b5039ef0b2a8bd2d6896bfe44ed.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "7b795ed2ed553e67e35237f25fb91e01e823ccb47b51aec4a43f2225da270a4f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bceb3c184da3eb644a1e6be126d893b197aeb362b3cb166b403bccaf96f6826a", "generation_seconds_batch": 21.195637193974108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7344747185707092, "target_z": 0.7812499965079963, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.6236240863800049, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7b795ed2ed553e67e35237f25fb91e01e823ccb47b51aec4a43f2225da270a4f.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "7ba6279a1d0aee5955d0a67d000aede36cf1d5fc594f5b25fd02a4530fc851a1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "81e6444b8df5a4e9639018b4739c55db3c2f9ea7abd2f74f4f72404c9b22cd20", "generation_seconds_batch": 21.00758527801372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9720057249069214, "target_z": 1.226562481666981, "ce": 6.09409761428833, "genuineness": 4.607763767242432, "identity": 0.6220021843910217, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7ba6279a1d0aee5955d0a67d000aede36cf1d5fc594f5b25fd02a4530fc851a1.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "7bc73f3e011bd2f1d04b177d61001ec4eb4e0b6d7e98d12b2bfa4a8dbe70ff9e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "064540e677414f68fcae87664d08e159c538486b8c1ec5642b7471c18454e88e", "generation_seconds_batch": 20.145719173015095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12814554572105408, "target_z": -0.35546876265851324, "ce": 6.086154937744141, "genuineness": 4.545003414154053, "identity": 0.6638416647911072, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7bc73f3e011bd2f1d04b177d61001ec4eb4e0b6d7e98d12b2bfa4a8dbe70ff9e.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "7bcbcdb37a17d2bffffa5a0af58d551dcfc926529d6ab3afe036955bc16b6320", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54c9c736421e59e0f47f0626c46b1f7255cdbaf6602a05cf278f79a698fb6c6e", "generation_seconds_batch": 22.024176714010537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18336109817028046, "target_z": -0.2519531195437443, "ce": 5.998785495758057, "genuineness": 4.20609712600708, "identity": 0.3376142978668213, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7bcbcdb37a17d2bffffa5a0af58d551dcfc926529d6ab3afe036955bc16b6320.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "7be3e92e31650760d622c4e8c6f7c98304fe7c64a82ac8001aa2ed3fb7439140", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d40a06edec450f5869bd5a69da5dd027d231a6d689851d6061493afc3c440914", "generation_seconds_batch": 22.55171699798666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8969959020614624, "target_z": 1.0859374275409244, "ce": 6.086154937744141, "genuineness": 4.469690799713135, "identity": 0.4425256550312042, "hypothesis": "I found the little box. It was b 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/emo_Confusion/7be3e92e31650760d622c4e8c6f7c98304fe7c64a82ac8001aa2ed3fb7439140.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "7be5bc0b7ab4e6c163038dccd08436de4b2f8c687217b76be82d6a27f2594361", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "487662e86894bc2ba00d4444870a1ed4b1e23f7be3385b5232bf92b3fdceefe0", "generation_seconds_batch": 20.193781876005232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12085293233394623, "target_z": -0.3691406274007525, "ce": 6.205295085906982, "genuineness": 4.645420074462891, "identity": 0.7314096689224243, "hypothesis": "La porte étant 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/emo_Confusion/7be5bc0b7ab4e6c163038dccd08436de4b2f8c687217b76be82d6a27f2594361.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "7c454d24d302ade949f091b0672715daacc068a5cad8c475912124a57908cda2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4c0e9e05b33cd66e91460744fd85fd0c61160edc9335673e31a736d59ceecdb7", "generation_seconds_batch": 19.561434058938175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13231275975704193, "target_z": -0.3476562565475068, "ce": 6.2767791748046875, "genuineness": 4.532451152801514, "identity": 0.3572803735733032, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7c454d24d302ade949f091b0672715daacc068a5cad8c475912124a57908cda2.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "7d3b5250dc87015693ab3d39a706e630b648400debf2f3d6495fe24a555724cd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1d25de1f25fde97a72942d28b612d7f674fa508c719e49d801adb6d1595857e4", "generation_seconds_batch": 21.750097501906566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1470285654067993, "target_z": 1.5546874589689572, "ce": 5.911416053771973, "genuineness": 4.13078498840332, "identity": 0.5076856017112732, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7d3b5250dc87015693ab3d39a706e630b648400debf2f3d6495fe24a555724cd.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "7d69085a90ab561327e4b8b55ebc4f109758e0fc89bbadb3871a48a042b954ea", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c11a8e48979dec842df78355f55027c348dd44a9d8c73f25816a9e38beff412d", "generation_seconds_batch": 20.525321257999167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19013281166553497, "target_z": -0.23925781457337716, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.2837550640106201, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/7d69085a90ab561327e4b8b55ebc4f109758e0fc89bbadb3871a48a042b954ea.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "7db20ebb6493b9676b686324e40822fbaf601647771974df3b7566bad76276ee", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "29d16ce49a8cdce6eb2e63cf886ad7287956c800b4b357cc7b9f0105c8934f6f", "generation_seconds_batch": 19.561434058938175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8136516809463501, "target_z": 0.9296874170649134, "ce": 6.157639026641846, "genuineness": 4.645420074462891, "identity": 0.4799996018409729, "hypothesis": "I found that 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/emo_Confusion/7db20ebb6493b9676b686324e40822fbaf601647771974df3b7566bad76276ee.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": "7f661bc783eef87e69ce788f5e4d47ea9e85b3475c084dc438ab2e60b8e0c298", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cca253eced9f9ad59557b78caf1cc274d3660eadc0c147206bd3a98cc3027165", "generation_seconds_batch": 16.24333581002429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6011239886283875, "target_z": 0.5312500244440256, "ce": 5.720792293548584, "genuineness": 4.532451152801514, "identity": 0.6979086995124817, "hypothesis": "Les clés sont sur la table de la cuisine J'ai regardé dans le tiroir, près de la fenêtre. Prends-les quand tu partiras.", "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/emo_Confusion/7f661bc783eef87e69ce788f5e4d47ea9e85b3475c084dc438ab2e60b8e0c298.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3dd1ff082448c93b2afcc0db87cc38cba13646a32cc876bf4aa6939746fe6a80", "generation_seconds_batch": 19.701358182006516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10210049152374268, "target_z": -0.40429686299623746, "ce": 6.356205940246582, "genuineness": 4.3127899169921875, "identity": 0.3810352683067322, "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/emo_Confusion/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "800628e1b2fa1ccc9907c466ca9fd23ba983704a8010335193dd8c08728181c8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "398501ae0f1cf4a19996c2bb85ba222e8e1a448b79a8377c7b84898483d74308", "generation_seconds_batch": 20.624740824918263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5512480735778809, "target_z": 2.3125000768240804, "ce": 6.181467056274414, "genuineness": 4.519899368286133, "identity": 0.49131786823272705, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/800628e1b2fa1ccc9907c466ca9fd23ba983704a8010335193dd8c08728181c8.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "801faf6e9d322363825bde17e9562b026fc0d1da80b29f59c47f95526c8a4ddf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6a389dc11d994beb2514e6803cd75e9883e87d4c6573049c400789e35d67726a", "generation_seconds_batch": 22.386052491958253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2013328075408936, "target_z": 3.531249800955792, "ce": 5.855817794799805, "genuineness": 4.193545341491699, "identity": 0.23728665709495544, "hypothesis": "De deur staat nog open. Ik heb je Jas op de stoel golgelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/801faf6e9d322363825bde17e9562b026fc0d1da80b29f59c47f95526c8a4ddf.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": "804e5f36e42d234139737e9e82968c5492bf7b87d0225725f306b2cf9a54ef0b", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d635d3f5b2c8dbc5679a4f5c9fa245ee937ba483bac00701c1e2a793a1994e3a", "generation_seconds_batch": 22.580438681994565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.234670639038086, "target_z": 3.5937500733320764, "ce": 5.613565921783447, "genuineness": 4.20609712600708, "identity": 0.37419646978378296, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/804e5f36e42d234139737e9e82968c5492bf7b87d0225725f306b2cf9a54ef0b.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "808698aa3bf1636a2b701c1ca23da62aed4052684268b988621d859856ed95e6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a500c5c9c4eb42e2dfa09cab4d3c111164851789f58da02ac593a76fcc691f40", "generation_seconds_batch": 15.671040626009926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0970220565795898, "target_z": 1.4609374973809972, "ce": 5.7684478759765625, "genuineness": 4.595211505889893, "identity": 0.6656027436256409, "hypothesis": "la porte est encore usierte. 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/emo_Confusion/808698aa3bf1636a2b701c1ca23da62aed4052684268b988621d859856ed95e6.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "80bc7bcc8f3d99912d6d10e7b2a8f1be9c32019b3aa1f4e842575e12a99f1674", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6cb4f63727b91797ec6af27bcc7806882e5005d269060b793abb899856715f98", "generation_seconds_batch": 16.033721885061823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.301215410232544, "target_z": 1.8437500453960474, "ce": 5.617537498474121, "genuineness": 4.444586753845215, "identity": 0.7823705673217773, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/80bc7bcc8f3d99912d6d10e7b2a8f1be9c32019b3aa1f4e842575e12a99f1674.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "d5e2885314b3c5a6ca53e8400008769c475d4ae43db90d1f6107fc4e79a3ec48", "generation_seconds_batch": 15.820263803005219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05938658118247986, "target_z": -0.4843749877779872, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.7054474353790283, "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/emo_Confusion/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "810e1efbacfad2a895ab0f34195e1bd12122c5facde8ed0296bcf327226cf415", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e8840dbcb0e3157614822a6b9c6e8dfec75e26f75164601fd261af589a9d9b54", "generation_seconds_batch": 20.502132874098606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6094583868980408, "target_z": 0.5468749807939799, "ce": 6.062326908111572, "genuineness": 4.7834930419921875, "identity": 0.6255124807357788, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/810e1efbacfad2a895ab0f34195e1bd12122c5facde8ed0296bcf327226cf415.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "813553e04c80843cad4449f95e34406d68e1b8b3eccf72ce4f0847fb2900587d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "816b8dd8131e4cf006be94b04e1456f184106ae3ae5df89acb78fb41b5e66998", "generation_seconds_batch": 21.073654906940646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12814554572105408, "target_z": -0.35546876265851324, "ce": 5.585766792297363, "genuineness": 4.256305694580078, "identity": 0.709414005279541, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "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/emo_Confusion/813553e04c80843cad4449f95e34406d68e1b8b3eccf72ce4f0847fb2900587d.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "8183c94692156a684663abc0e292c507a8f93a6bbb57c911239bce32a2cf4fa9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6dc13feb053387e0acc8b6a31d03a23a9c852d4faa1de7f7d520bd7e5e8dc6b5", "generation_seconds_batch": 19.504626156995073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6012545824050903, "target_z": 2.40625003841204, "ce": 5.927301406860352, "genuineness": 4.149612903594971, "identity": 0.682623028755188, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8183c94692156a684663abc0e292c507a8f93a6bbb57c911239bce32a2cf4fa9.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "819c5254a3f1419ad37d81b833c692b898f966ec321c2aa9bc76db24ad0904de", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b37611b69d902d96376264c2516bd38ad221b14447c6349b8b16ef5edaa08a6b", "generation_seconds_batch": 20.714532652986236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9970090389251709, "target_z": 1.2734375742050774, "ce": 5.784333229064941, "genuineness": 4.469690799713135, "identity": 0.6326137185096741, "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/emo_Confusion/819c5254a3f1419ad37d81b833c692b898f966ec321c2aa9bc76db24ad0904de.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "8242b425da91f12729b828ef1ba79898f9132deee83b2b91e5aa450866bd9046", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ceb6f0ffc225be62aa5f89fcef1739d6211f7d7dfa3dc2d464a5341d11071d0b", "generation_seconds_batch": 27.581000043894164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1562742292881012, "target_z": -0.30273436736124204, "ce": 6.268836498260498, "genuineness": 4.595211505889893, "identity": 0.43437349796295166, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8242b425da91f12729b828ef1ba79898f9132deee83b2b91e5aa450866bd9046.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "8310e1cf105c5992a3c83caa0e1db74d7bb767c3f84647ee80c3ca7b09c4cf2d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1d1e41a17d467682be6dc5cd11a20d5183e88a2df4458d72ca70ace7a4bc6d4c", "generation_seconds_batch": 15.389283393044025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.126192569732666, "target_z": 1.515625068094071, "ce": 6.086154937744141, "genuineness": 4.319066047668457, "identity": 0.786485493183136, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8310e1cf105c5992a3c83caa0e1db74d7bb767c3f84647ee80c3ca7b09c4cf2d.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "83daaa7ac660092c37ecae412de1b252c04c4e2007a24d50bab5f0f4306096d3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2e1ec57be4c0c35acc494fad587bc4e76e71fdf220997f877b21244d4d35b22f", "generation_seconds_batch": 19.111097343964502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7969828844070435, "target_z": 0.8984375043650046, "ce": 5.982900142669678, "genuineness": 4.281409740447998, "identity": 0.5998835563659668, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/83daaa7ac660092c37ecae412de1b252c04c4e2007a24d50bab5f0f4306096d3.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "8410de2642e002f08731eae1b08a2fff3e9fe8c942d0ecd9704938fcf67c89c0", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "782054013089b1a3217a8bbe20dc2846212641354a2de6d719101e53fcf4adb5", "generation_seconds_batch": 14.635950949974358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 5.788304805755615, "genuineness": 4.105680465698242, "identity": 0.7706798315048218, "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/emo_Confusion/8410de2642e002f08731eae1b08a2fff3e9fe8c942d0ecd9704938fcf67c89c0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "8432914b59be8a28c6d681574178d0b52a5802b7ae4134055f6c7fa822077ef6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b766a6df70b9ffe90958e6cd38570482bf8c9882cf5d02711dc383f2a6f25b6c", "generation_seconds_batch": 21.177059186040424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07293000817298889, "target_z": -0.458984377837253, "ce": 6.022613525390625, "genuineness": 4.13078498840332, "identity": 0.6954057812690735, "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/emo_Confusion/8432914b59be8a28c6d681574178d0b52a5802b7ae4134055f6c7fa822077ef6.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "8478ec4f8b1249a1ff529e598db4afe7e24c2c9a41a7a44b3a7e4810c8079566", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd1442ffc839feef3600336ea6a1025c3f70da264b31428e7573307f99240c83", "generation_seconds_batch": 20.465764509979635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0843898355960846, "target_z": -0.4375000069840073, "ce": 6.1894097328186035, "genuineness": 4.256305694580078, "identity": 0.44065922498703003, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8478ec4f8b1249a1ff529e598db4afe7e24c2c9a41a7a44b3a7e4810c8079566.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "8497666ec7c831eae26512099d264ae5e834d9c42f5ac89a7aedd4a6163e02b7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "27770649dda20ff039812a187de2792dad041a8a0e33b935878d374f58d51067", "generation_seconds_batch": 16.170916605042294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1513264179229736, "target_z": 3.4375000628560657, "ce": 5.315715789794922, "genuineness": 3.8546390533447266, "identity": 0.38346436619758606, "hypothesis": "Ik 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/emo_Confusion/8497666ec7c831eae26512099d264ae5e834d9c42f5ac89a7aedd4a6163e02b7.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "84e89b3867cccd182cc73e945e6063ed8db35e2548b702e9a24893a6c2001552", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a03dff545ce6d596189e2d767fca49f9e08bcad81054205773152a9c8d34277e", "generation_seconds_batch": 16.31303141801618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2553761005401611, "target_z": 1.7578125619830647, "ce": 5.71682071685791, "genuineness": 4.7458367347717285, "identity": 0.7283421754837036, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/84e89b3867cccd182cc73e945e6063ed8db35e2548b702e9a24893a6c2001552.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "85060eb1013cee0f5ec5998638eedc81976a17fb37f2d20769739b865f08dc10", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3264fcda05f306798cdc5e337cceafb234860ace8de50d63a029e60697d733e5", "generation_seconds_batch": 20.364825114025734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11981113255023956, "target_z": -0.3710937469444968, "ce": 5.959072113037109, "genuineness": 4.758388519287109, "identity": 0.7952823042869568, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/85060eb1013cee0f5ec5998638eedc81976a17fb37f2d20769739b865f08dc10.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "85082f1f8a277565966163eff09bbfab40692ebc79c7d10b52966dd3302c8401", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2de51d1a61ff87c8b10c19cf1864ea1767cab6c69557e28768744025f9435538", "generation_seconds_batch": 20.96034846501425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0053434371948242, "target_z": 1.2890625305550318, "ce": 5.403085231781006, "genuineness": 4.482243061065674, "identity": 0.6823244094848633, "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/emo_Confusion/85082f1f8a277565966163eff09bbfab40692ebc79c7d10b52966dd3302c8401.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "850b0363812f8d0d9505c549082222ba186edc1aca69677ad0c07977afd2778f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f9de08def491a46ae02f6bbe10aeb3293b16daa654b43e4c20ed35b3ae8974ad", "generation_seconds_batch": 20.763286706991494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.011463642120361328, "target_z": -0.5742187661505168, "ce": 5.95112943649292, "genuineness": 4.7458367347717285, "identity": 0.6583410501480103, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/850b0363812f8d0d9505c549082222ba186edc1aca69677ad0c07977afd2778f.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "852a184fb4e0ec7f811331118ffe3d2f62c512c3c4db86e903a5e5ef8b97ba18", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "186f9d155003aad1b17183bf5bd2264a3342145b9c6cfe6e34c80549cda33fbf", "generation_seconds_batch": 15.536983537953347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13647997379302979, "target_z": -0.33984375043650045, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.6114705801010132, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/852a184fb4e0ec7f811331118ffe3d2f62c512c3c4db86e903a5e5ef8b97ba18.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "854c631ae58bdaccaf67ee1c28fd9e693a7abae74bcd5f45a42fac820231c5fa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0bd4308d53f82f9c4d8efae4175d270a244e338ba6c34a8a7f79fbc6c0e24853", "generation_seconds_batch": 15.814881471917033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4193294048309326, "target_z": 0.19042968629962376, "ce": 5.704906940460205, "genuineness": 4.482243061065674, "identity": 0.8615988492965698, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/854c631ae58bdaccaf67ee1c28fd9e693a7abae74bcd5f45a42fac820231c5fa.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "85c43de91e63681189840113919551c492fdc23e81b30f53a956759096720cff", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7acda23e486ce85f3b89d1b436718bf7dac470084ef517f432df72bc2ec249a9", "generation_seconds_batch": 21.635982164996676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.4912309646606445, "genuineness": 4.494794845581055, "identity": 0.4837460517883301, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/85c43de91e63681189840113919551c492fdc23e81b30f53a956759096720cff.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "85dd39c25a4288f40336e51a8bce1e64ce13efc66680a14af97bb43255dde9b9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "48bcdba5e45b4495acd33ffc4f0dc5f221dacd4428bcdd9c256e3f026d781cdd", "generation_seconds_batch": 19.32685847999528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 5.967014789581299, "genuineness": 4.6705241203308105, "identity": 0.7042820453643799, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/85dd39c25a4288f40336e51a8bce1e64ce13efc66680a14af97bb43255dde9b9.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "86965c0f2365261d63f1a02ab6be9ffc7b01d1807320d696b20b576085c9fcf0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c9dde34dd78391a7a581060b24a86ba314d74109f5f4b189628752262da019bc", "generation_seconds_batch": 21.47277455602307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16356684267520905, "target_z": -0.28906250261900274, "ce": 6.173524379730225, "genuineness": 4.262581825256348, "identity": 0.34735846519470215, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/86965c0f2365261d63f1a02ab6be9ffc7b01d1807320d696b20b576085c9fcf0.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "86c2cf54d805d6ab43f00da623746b14d48ef295ceb55d784db8a40256e072fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8960536457d7e6ec5f40de17366d54f7190663498ca8e05b7ce124d051f33af8", "generation_seconds_batch": 22.0632179339882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.151195764541626, "target_z": 1.5624999371439343, "ce": 5.589737892150879, "genuineness": 4.532451152801514, "identity": 0.7220338582992554, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/86c2cf54d805d6ab43f00da623746b14d48ef295ceb55d784db8a40256e072fe.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "87741e4e43d476a739d8011e3814dc649684a3d209fa88e8d32b338c0f1ce4ae", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "dd376832c58916a6adfa3e3edb27a2073b5bd27b15616857079053acd3e1f23e", "generation_seconds_batch": 22.55171699798666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7073878645896912, "target_z": 0.7304687766265278, "ce": 6.086154937744141, "genuineness": 4.287685871124268, "identity": 0.37876835465431213, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/87741e4e43d476a739d8011e3814dc649684a3d209fa88e8d32b338c0f1ce4ae.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "87742d9a2dac07b192e236d0bed8b4e20ccac8ac09969cc39be63ba72c2f7c16", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d6add0c6fc4860a77bef6183fdbca3c2c30e15158e3bbfc9cd345b1540a20760", "generation_seconds_batch": 21.86774013703689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.767943024635315, "target_z": 2.718750059364062, "ce": 6.022613525390625, "genuineness": 4.18099308013916, "identity": 0.7979537844657898, "hypothesis": "Ik 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/emo_Confusion/87742d9a2dac07b192e236d0bed8b4e20ccac8ac09969cc39be63ba72c2f7c16.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "877d3043df315d4982a703d2a9897de7ee5488850becad54bd7940e63532e30a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8fd36d5122ac467ca5e2436b6fa31e09b1018465f40862c764b1152de4e52683", "generation_seconds_batch": 21.53320384095423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8844943046569824, "target_z": 1.0624999930159926, "ce": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.5327110290527344, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/877d3043df315d4982a703d2a9897de7ee5488850becad54bd7940e63532e30a.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "87817c898a037b9b4c5f107209906fea4e301aedc40d60aecdad315a87f30e09", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7d5081cd5b8bd2252348fc6a0b8691502710961ba599c4e27e6209b406f3adf7", "generation_seconds_batch": 15.782314590062015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.013547241687774658, "target_z": -0.5703125270630283, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.7765064239501953, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/87817c898a037b9b4c5f107209906fea4e301aedc40d60aecdad315a87f30e09.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "87a95600a894dec69bdf4d57b9aaedb2c635b0ee49183389467d112876aa9a9f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "f298226dc98bc68ceda7aca3e9b0386f1acb385fa0a7366b6303b8be4dc7aabf", "generation_seconds_batch": 15.0956993320724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4906928837299347, "target_z": 0.3242187382144877, "ce": 5.816103935241699, "genuineness": 4.758388519287109, "identity": 0.7333924770355225, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/87a95600a894dec69bdf4d57b9aaedb2c635b0ee49183389467d112876aa9a9f.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "88080987c02cb648fa6ceb489ca3f8afec708540f0494dd741f16baa15d68938", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7f916e666b04bd73af52a3254ac3636a751d1e930afe0328a218edc562c42169", "generation_seconds_batch": 20.218488796963356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4263622760772705, "target_z": 3.953124963333962, "ce": 5.7128496170043945, "genuineness": 4.168440818786621, "identity": 0.09913937747478485, "hypothesis": "De deur 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/emo_Confusion/88080987c02cb648fa6ceb489ca3f8afec708540f0494dd741f16baa15d68938.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "887c32f314ffb63e8b3516c637985d33f0927af592eb9e80165798bae2ba8853", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fb18f20e30b43574e153d8ebd8d1397f58b8bef7a38c517183ee97f2e12bc0fe", "generation_seconds_batch": 21.735346120898612, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11981113255023956, "target_z": -0.3710937469444968, "ce": 6.300607204437256, "genuineness": 4.482243061065674, "identity": 0.3647887110710144, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/887c32f314ffb63e8b3516c637985d33f0927af592eb9e80165798bae2ba8853.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "899c8e738cd6e4f249233b44bd9571e156a369c911852d23584e85522a7f7092", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "87b8b75e7e5dd80033bcb5ec599012f41111b64dbc4bce4bfc2816bc9e9f3370", "generation_seconds_batch": 14.954697004053742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2178711891174316, "target_z": 1.6875000349200364, "ce": 5.9431867599487305, "genuineness": 4.268857479095459, "identity": 0.5542754530906677, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/899c8e738cd6e4f249233b44bd9571e156a369c911852d23584e85522a7f7092.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "8a2171de7764ba396dcc99df8b39a431fd36fdbe9ac68bb80abd7124ba8ac3d4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d598e7b01dff9d301ad8cd32de346e9673a514d09feb39b8645fe49c4e7313bd", "generation_seconds_batch": 19.928308811970055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12814554572105408, "target_z": -0.35546876265851324, "ce": 6.356205940246582, "genuineness": 4.519899368286133, "identity": 0.35006922483444214, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8a2171de7764ba396dcc99df8b39a431fd36fdbe9ac68bb80abd7124ba8ac3d4.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "8a8c840a50f53860d60383358f765501dd1d8802956166503caa2d901594611f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9665f1a18d522705bce51324c1de80c3003f2f8ff38fefcc7ca569bb2ce0a304", "generation_seconds_batch": 21.691767125972547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5095759630203247, "target_z": 2.234375071586075, "ce": 6.117925643920898, "genuineness": 4.287685871124268, "identity": 0.726749062538147, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8a8c840a50f53860d60383358f765501dd1d8802956166503caa2d901594611f.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "8a9598f0bee6446dd82d4523a512c75ef9f498913f3c42b3d4b93addafbf565d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e2f42b87154c110554bcf2c64377ccb2e92d47be88090928c51ce347bf116eb3", "generation_seconds_batch": 20.761759391985834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 5.9431867599487305, "genuineness": 4.7332844734191895, "identity": 0.7291126251220703, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8a9598f0bee6446dd82d4523a512c75ef9f498913f3c42b3d4b93addafbf565d.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "8ac7905884194c92a41836c31b68636af59cfaeee77b0ad6fabb43a43dee3c7b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a77851e715515f8008d4c1b5e02ae99cd32afeb7d2e3d5807d6a70bd0677e0a8", "generation_seconds_batch": 15.74047174397856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.173524379730225, "genuineness": 4.369274139404297, "identity": 0.5446798205375671, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8ac7905884194c92a41836c31b68636af59cfaeee77b0ad6fabb43a43dee3c7b.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "8b21e9e531dd110b611b2adef417fa21349ce71a1bb930787a4836aa42c9a0a4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "52a0abd7613392fb18b7ffbeff3a184174caf1dc5ec42bf4cd064649787c8b59", "generation_seconds_batch": 20.111027395003475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0220122337341309, "target_z": 1.3203124432549407, "ce": 5.788304805755615, "genuineness": 4.457139015197754, "identity": 0.5592818260192871, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8b21e9e531dd110b611b2adef417fa21349ce71a1bb930787a4836aa42c9a0a4.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": "8bcdda69e0e6db57c06059256df90644666836f373bf705f6fdf75272be37ce2", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1a3bcbae214bf7b7cbe65c7d7da62c34d376719689ed63a8e92b233f6145246e", "generation_seconds_batch": 20.5698393820785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8636583089828491, "target_z": 1.0234376021411067, "ce": 5.959072113037109, "genuineness": 4.482243061065674, "identity": 0.7407152056694031, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8bcdda69e0e6db57c06059256df90644666836f373bf705f6fdf75272be37ce2.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "8cefb8ec233b5b61992536bc7a91a302d7c35be437b1761483f19dd9b80ac855", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b67587f59c67fff3349cfde25fbf5fc8fb7e08e363a098c2f375d832fc2a7a52", "generation_seconds_batch": 20.78185544698499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8594909906387329, "target_z": 1.015624900477896, "ce": 6.038498878479004, "genuineness": 4.356721878051758, "identity": 0.8001024723052979, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8cefb8ec233b5b61992536bc7a91a302d7c35be437b1761483f19dd9b80ac855.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "8e1b0ab33dc94c222ba0d1e6c7ddeea1ed8f28a9f525962c80e8da78e7bf3153", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "545eced2c7de82c9acc920a4982edae715b2b628932c578dd5bd30f1c8934568", "generation_seconds_batch": 20.82704621704761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6012545824050903, "target_z": 2.40625003841204, "ce": 5.855817794799805, "genuineness": 4.645420074462891, "identity": 0.6007828712463379, "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/emo_Confusion/8e1b0ab33dc94c222ba0d1e6c7ddeea1ed8f28a9f525962c80e8da78e7bf3153.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "8ecd7c7cb42d7867252f861e80ef85dbbb697c6c1b7cf6c39560880415f357de", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10afae93cf6498a69a1552c88729486badacd2deee2d5905b10aa8384ba707be", "generation_seconds_batch": 16.22359822492581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2635799050331116, "target_z": -0.10156249563499545, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.4533422589302063, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8ecd7c7cb42d7867252f861e80ef85dbbb697c6c1b7cf6c39560880415f357de.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "8ed7805ae417c0ac1baaf6c8e4491ddefb2a30a64ae0c332358166d631e819c6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "148bb4d73dbd550206689cd6f0ff047216f7d4fdbed732b55e2b4357dd77b333", "generation_seconds_batch": 20.664601166034117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40370237827301025, "target_z": 0.16113283727140087, "ce": 5.982900142669678, "genuineness": 4.6203155517578125, "identity": 0.5850685834884644, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8ed7805ae417c0ac1baaf6c8e4491ddefb2a30a64ae0c332358166d631e819c6.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": "8f3abf9b98e93e31023ddf4f00c84019ddbcc5242e19c05ac0ab39d3e15313ff", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d294530b13542d5e4d952b4392cd8a3b8331e85304afcb6e245a4e38c95272fa", "generation_seconds_batch": 15.319388268049806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1929984092712402, "target_z": 3.5156248446058376, "ce": 5.470598220825195, "genuineness": 4.13078498840332, "identity": 0.6251518726348877, "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/emo_Confusion/8f3abf9b98e93e31023ddf4f00c84019ddbcc5242e19c05ac0ab39d3e15313ff.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": "8f3fc7a481cf263cb67fadca258c2df801e34c3b9b827763c61b1999e44e6381", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The road turns left behind the bridge.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc23b126f11902d93a64a1e3cefb27974df9e1050a919fa702e753de46e2fafd", "generation_seconds_batch": 15.828915882972069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14273078739643097, "target_z": -0.3281250052380055, "ce": 6.0146708488464355, "genuineness": 4.243753433227539, "identity": 0.2416190803050995, "hypothesis": "We row turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "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/emo_Confusion/8f3fc7a481cf263cb67fadca258c2df801e34c3b9b827763c61b1999e44e6381.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "8f511ceae200d113dff1b2fb57a0e4c75d94629ff599a24211f044a1b88008fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a6a40eeab87308a0ec20b255c93a226522f88aaeded890ed25665f487706fde5", "generation_seconds_batch": 24.712741278926842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8762905597686768, "target_z": 2.9218751623781696, "ce": 5.863759994506836, "genuineness": 4.231201171875, "identity": 0.4589846134185791, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8f511ceae200d113dff1b2fb57a0e4c75d94629ff599a24211f044a1b88008fe.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "8f5f957f31adb499db47db43a0c61a7894703ffd7ad5bc15892356b4829ed175", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5064676334e546ba892cf054b61785e4f9e5a83215348c543db6bf48e1752b8", "generation_seconds_batch": 21.902440170990303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1760684698820114, "target_z": -0.2656250122220128, "ce": 6.022613525390625, "genuineness": 4.7332844734191895, "identity": 0.6611407995223999, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8f5f957f31adb499db47db43a0c61a7894703ffd7ad5bc15892356b4829ed175.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "8fa74a1e2a70ce1922bc25b983d1d63ef035bbfff5dfb31559f4c04041d41d64", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "21783868fd16665ae30cd290a2b30bc1161ca609ca4624fb9df524a9c580064c", "generation_seconds_batch": 14.91636687098071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4679038524627686, "target_z": 2.1562500663480693, "ce": 5.669164657592773, "genuineness": 4.306513786315918, "identity": 0.7111506462097168, "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/emo_Confusion/8fa74a1e2a70ce1922bc25b983d1d63ef035bbfff5dfb31559f4c04041d41d64.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "8fb4056a105c081b2951954c89bd5d8cfd46ac2e8a3e3b493aa20916c636d065", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a54e4b932be6ad0215ee66ecc27ca4c85fa93eef3d65aa5f0689f2d9415367f8", "generation_seconds_batch": 14.700863317935728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06355378031730652, "target_z": -0.47656250960301005, "ce": 6.268836498260498, "genuineness": 4.137060642242432, "identity": 0.3091133236885071, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/8fb4056a105c081b2951954c89bd5d8cfd46ac2e8a3e3b493aa20916c636d065.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "90601808ea3a3110c3331e3bae90f10a213dca363b419607382ecb6255765fa6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a69cf88dbb1decac189fbcea0696a5b750cef9c88d5d05eb2790192b9e6c3504", "generation_seconds_batch": 18.237061318941414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36020711064338684, "target_z": 0.07958985886382829, "ce": 5.8875885009765625, "genuineness": 4.331617832183838, "identity": 0.24497577548027039, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/90601808ea3a3110c3331e3bae90f10a213dca363b419607382ecb6255765fa6.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "9088420836523a60efe424b6c7a2b4c0c5834f72ec788acea2048b77044968c5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13c0a45cb6c07a859656207e49d20049e041b6cf45f3099eba031932c536bcd6", "generation_seconds_batch": 20.525321257999167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08230623602867126, "target_z": -0.4414062460714959, "ce": 6.268836498260498, "genuineness": 4.375550270080566, "identity": 0.3893193006515503, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9088420836523a60efe424b6c7a2b4c0c5834f72ec788acea2048b77044968c5.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "90913b760188fb147ff7f7a069a9c978b9a5470aba66af50fbb37df9cc1a9e30", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ac0f620a7fb8a364b098ee40f1281250baf7447082e0a5a0e834784b21ddc3a", "generation_seconds_batch": 14.700863317935728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07397180795669556, "target_z": -0.45703125829350866, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.2648151218891144, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/90913b760188fb147ff7f7a069a9c978b9a5470aba66af50fbb37df9cc1a9e30.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "90cafcb6416c39f2c4af9deebcbc78df1dfdca7805d6fb53426cf39aad351fd7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "853e9eb3a212cbee5dafe2398f8454522b64ca071d13e066da16352f96a8049a", "generation_seconds_batch": 20.305700591066852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.6973636746406555, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/90cafcb6416c39f2c4af9deebcbc78df1dfdca7805d6fb53426cf39aad351fd7.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "90cb904cc00df1557649df49e089d092c4a58e0cca1e266bc8b0a4561d44e015", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2f48bc8439fd08bd25d150cf6ae2737af5c602eaf3abc66655e77749007575b9", "generation_seconds_batch": 21.53320384095423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7386419177055359, "target_z": 0.7890624746829735, "ce": 5.871703147888184, "genuineness": 4.193545341491699, "identity": 0.4861600995063782, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/90cb904cc00df1557649df49e089d092c4a58e0cca1e266bc8b0a4561d44e015.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "9132b7f6728bd9f40962ef0f012f26a912d2d9009a7d34505282a6aa36a60d93", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "aa2b8aa4980b10e08ca7b6a17e2f693b36214ece1e084bcab934710b45d4ff1a", "generation_seconds_batch": 16.43362798599992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.976172924041748, "target_z": 1.234374959841958, "ce": 5.871703147888184, "genuineness": 4.406930446624756, "identity": 0.42083126306533813, "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/emo_Confusion/9132b7f6728bd9f40962ef0f012f26a912d2d9009a7d34505282a6aa36a60d93.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "91b38b168351b70797a7551ed11a2106809561b87f33d6dc038e851344fdb7db", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a26fc95c9f60a8d3b1951950766e5598cbfc44376a3c8ce2ab7af76f3eafe8fc", "generation_seconds_batch": 23.01439790194854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.492907166481018, "target_z": 2.203125158886166, "ce": 5.307773113250732, "genuineness": 4.243753433227539, "identity": 0.23011401295661926, "hypothesis": "J'ai 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/emo_Confusion/91b38b168351b70797a7551ed11a2106809561b87f33d6dc038e851344fdb7db.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": "91b7e462719585a3ccbcdc0abf1074ec9ad0d71e0371eae3d320cf8b4e4790d0", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3be8db308947ce2c0b1c5e006d5b09d150fdf6c31baf6fd171acd0a0c3843d44", "generation_seconds_batch": 14.641263417084701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8596216440200806, "target_z": 2.8906250261900275, "ce": 5.967014789581299, "genuineness": 4.507347106933594, "identity": 0.5556603670120239, "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/emo_Confusion/91b7e462719585a3ccbcdc0abf1074ec9ad0d71e0371eae3d320cf8b4e4790d0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "927ddbcec6b17f0ef87800f41d9067f74a60963a217353ecce9cec30ae9fd13f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e055686760301759efa9adad31b86cb385abf470e948e6bce55ca2bc49f20207", "generation_seconds_batch": 21.086008926969953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6262578964233398, "target_z": 2.453125130950137, "ce": 5.708878040313721, "genuineness": 4.507347106933594, "identity": 0.6025776267051697, "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/emo_Confusion/927ddbcec6b17f0ef87800f41d9067f74a60963a217353ecce9cec30ae9fd13f.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "927ee590a61de75d1a4f3a19e6587f20e58682d0a992cf8b2c117fcefb592de5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d3e64e718a698379a43a12e8a57cc1c535b73485757acbc7c754a3087af049ef", "generation_seconds_batch": 16.000706993043423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 5.824047088623047, "genuineness": 4.494794845581055, "identity": 0.7700172662734985, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/927ee590a61de75d1a4f3a19e6587f20e58682d0a992cf8b2c117fcefb592de5.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": "933650de911583850b49d6cc11bcb376572877ebf61b93608e0fe2ea6e1401b8", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85cb00054c4a3b7c7e4e489c8aa17ecfc880d7bc37a56751f3ad16075ccb5c04", "generation_seconds_batch": 13.613664992968552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8512872457504272, "target_z": 2.875000069840073, "ce": 5.661221981048584, "genuineness": 4.344170093536377, "identity": 0.5844255685806274, "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/emo_Confusion/933650de911583850b49d6cc11bcb376572877ebf61b93608e0fe2ea6e1401b8.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "937792c638c68aa744768616f10a2eb3b4fc04c51dbe0b8ebb0b90b918485686", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3afe0c6a1ce2e83613840903a86cc92dace7f87c15a856943aef6185add99f5f", "generation_seconds_batch": 15.70324643899221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8636583089828491, "target_z": 1.0234376021411067, "ce": 5.824047088623047, "genuineness": 4.557555198669434, "identity": 0.6766462326049805, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/937792c638c68aa744768616f10a2eb3b4fc04c51dbe0b8ebb0b90b918485686.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "93f37e245d313ac9fc62907486f7c80a9eb1a7cc2977b5ef3c3ccc84df0f3b9d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "349d8e230abf748e29e346777550d2872908fcd3f394a74da4ecaf3ba25dce22", "generation_seconds_batch": 19.666661383002065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6886354088783264, "target_z": 0.6953125130950136, "ce": 5.967014789581299, "genuineness": 4.632867813110352, "identity": 0.7650160789489746, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/93f37e245d313ac9fc62907486f7c80a9eb1a7cc2977b5ef3c3ccc84df0f3b9d.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "94b833dddc65155d8ddd39658762fe5ca4407d17fd2ba13942d1b90366ecc80b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f993a6076ef984618ccbf3c724c20fe54db6376ff156921253cf784f7fc1a0f7", "generation_seconds_batch": 21.139076616964303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21148976683616638, "target_z": -0.1992187521825023, "ce": 6.133810997009277, "genuineness": 4.7081804275512695, "identity": 0.679236888885498, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/94b833dddc65155d8ddd39658762fe5ca4407d17fd2ba13942d1b90366ecc80b.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "951241db1474616bb649e9d0e4e69eb88311a958737987f61d0f0577fa144bf6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a8296344be6599a8d52c9102afb05995e121fe378acc8a81482fa25dafcbe3d", "generation_seconds_batch": 19.705652939970605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1218947321176529, "target_z": -0.3671875078570082, "ce": 6.022613525390625, "genuineness": 4.319066047668457, "identity": 0.44252288341522217, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/951241db1474616bb649e9d0e4e69eb88311a958737987f61d0f0577fa144bf6.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": "9565dfd9d3f972df62f02cdc3af904732ff2077a09df78ecacb15449b1307b14", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec897ec05a8d6aee7f6ddfc234249aba91a910229413e03362c9619154388f24", "generation_seconds_batch": 21.95666484697722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.151195764541626, "target_z": 1.5624999371439343, "ce": 5.665193557739258, "genuineness": 4.293961524963379, "identity": 0.7385698556900024, "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/emo_Confusion/9565dfd9d3f972df62f02cdc3af904732ff2077a09df78ecacb15449b1307b14.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "95c22059eaabe09aaccc5b401cf4c2dc42e6264e390d7f481e5f17f7abc9d9fa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8538e25650d8c95eb46e59cae31cdf92f5a09445aaeb2e2f314c70349bab093d", "generation_seconds_batch": 16.02757140796166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7512742280960083, "target_z": 2.6875001466641533, "ce": 5.5659098625183105, "genuineness": 4.607763767242432, "identity": 0.6424025297164917, "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/emo_Confusion/95c22059eaabe09aaccc5b401cf4c2dc42e6264e390d7f481e5f17f7abc9d9fa.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "95d829d501ef65637b7ce011ae8741b6e7e9e051a588353068716685a57d4f8b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2bc990ef16afadf5ce399af911c2ec5058bdca4ce000ef28b85554a169e27aba", "generation_seconds_batch": 21.21281185396947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38000136613845825, "target_z": 0.11669924193908672, "ce": 6.141753673553467, "genuineness": 4.758388519287109, "identity": 0.7695456743240356, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/95d829d501ef65637b7ce011ae8741b6e7e9e051a588353068716685a57d4f8b.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "968ec77fcfccce0a6403a423b0a7b7f2aa9076a1003e42e67e149b493f02c3d1", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6ffe80b8f76ce7cb6f85b743d6b0692f30d30ffb88d5c73f767da9aad1b01169", "generation_seconds_batch": 17.45860060909763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07605542242527008, "target_z": -0.4531249912699909, "ce": 6.205295085906982, "genuineness": 4.394378662109375, "identity": 0.4590305685997009, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/968ec77fcfccce0a6403a423b0a7b7f2aa9076a1003e42e67e149b493f02c3d1.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "96ab1eeec7fe81af029fa3e8ec82fd1d9152e4300c1ebd22a3c4ec2c8350707d", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d26333493bf6474974ced216dd136c11daab52525fd211aab20e3f710c0a579e", "generation_seconds_batch": 13.717525087064132, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9136648178100586, "target_z": 1.1171875637290665, "ce": 5.974957466125488, "genuineness": 4.256305694580078, "identity": 0.561988115310669, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/96ab1eeec7fe81af029fa3e8ec82fd1d9152e4300c1ebd22a3c4ec2c8350707d.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "96b14b2ad60fd5bab985031821ca85b4ae8b7c4c84a91876dd5a2ad98f45d608", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b97ee74988649078225456eeb59b88489f1ff7ea92651acb5bdb5068b18b7071", "generation_seconds_batch": 15.683584069018252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9012937545776367, "target_z": 2.9687500314280326, "ce": 5.911416053771973, "genuineness": 4.250029563903809, "identity": 0.3426705598831177, "hypothesis": "Ik heb het kleine oosje 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/emo_Confusion/96b14b2ad60fd5bab985031821ca85b4ae8b7c4c84a91876dd5a2ad98f45d608.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "96c7a45c1309c4380a3dbcd896e2a35add3b3b64a1cac3e14472281dddf99cd8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7a734c368392fbca9c67c289718dc51cf7213c5c532dd92a5a296195bcc9327b", "generation_seconds_batch": 19.928308811970055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2182614803314209, "target_z": -0.18652344721213515, "ce": 6.1894097328186035, "genuineness": 4.595211505889893, "identity": 0.3793922960758209, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/96c7a45c1309c4380a3dbcd896e2a35add3b3b64a1cac3e14472281dddf99cd8.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "96e3691f938e34da376e5f250d7b5c18351c2177afa8ff4a2779f97dc9f22132", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9489a55aec60c3cb01d96422ac2ea63c691e958e1b95364b9b9422ee343a7da9", "generation_seconds_batch": 21.996295294957235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8346184492111206, "target_z": 2.843750157140164, "ce": 5.9431867599487305, "genuineness": 4.262581825256348, "identity": 0.6329153776168823, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/96e3691f938e34da376e5f250d7b5c18351c2177afa8ff4a2779f97dc9f22132.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": "971e8b11cb7a67dc62db21d48af3ec5fdacf6c24730bd6b9f5af4ff3065c37a9", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Escribí el número en el sobre.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddcb97aa76651ec2e483aeb48b6e5856a53628676bdb71645e5a65e9160daf0c", "generation_seconds_batch": 15.43189180700574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1470285654067993, "target_z": 1.5546874589689572, "ce": 5.824047088623047, "genuineness": 4.607763767242432, "identity": 0.7209830284118652, "hypothesis": "Escribí el número en esora, léelo 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/emo_Confusion/971e8b11cb7a67dc62db21d48af3ec5fdacf6c24730bd6b9f5af4ff3065c37a9.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "97382712a65edb1167fe16d393e7bcc4c8a54572c2b9f3a4203098aa267a3c3a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "13f981b0987812a6bce5fb83330bee62a0a56eb7fbed1e6baeb29ed26cdcb596", "generation_seconds_batch": 15.182318886974826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4095629453659058, "target_z": 2.046875148410155, "ce": 5.927301406860352, "genuineness": 4.519899368286133, "identity": 0.7664176821708679, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/97382712a65edb1167fe16d393e7bcc4c8a54572c2b9f3a4203098aa267a3c3a.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "977031a0b58c1b77ba83ca00a40722bfdfda87075ed91dc650bfc5b7acbce564", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f1811965942830ec3d91a43b8afb89399761670fc7691c1111edddce6d5c6726", "generation_seconds_batch": 24.909868221031502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.592920184135437, "target_z": 2.3906250820620856, "ce": 5.657250881195068, "genuineness": 4.080576419830322, "identity": 0.5938475131988525, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/977031a0b58c1b77ba83ca00a40722bfdfda87075ed91dc650bfc5b7acbce564.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "97f1ace85434f1e9a2d6142c501e6b855bf3ebb8e9cb481892d64cc3d6f5ab9c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c2beed7787c6c63853fb6efce58a58a6ef3d0e5e9344c28545ac26dce4b707a5", "generation_seconds_batch": 21.301815377897583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7012676000595093, "target_z": 2.59374996158796, "ce": 5.871703147888184, "genuineness": 4.344170093536377, "identity": 0.5193042159080505, "hypothesis": "Ik heb het een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen.", "wer": 0.16666666666666666, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/97f1ace85434f1e9a2d6142c501e6b855bf3ebb8e9cb481892d64cc3d6f5ab9c.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "993bcb3a260cd321e7f9619961dc45e2e74df7e5e7104ce3387632ad534f5822", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bc2cb0bda0b09121d76ab087b6f8e40af33f645529a5efbd8d5eb0a89c4cebbb", "generation_seconds_batch": 22.34608659008518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00103798508644104, "target_z": -0.5976562565475069, "ce": 6.38003396987915, "genuineness": 4.557555198669434, "identity": 0.6648708581924438, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/993bcb3a260cd321e7f9619961dc45e2e74df7e5e7104ce3387632ad534f5822.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": "997f7c17c63f755dfffde3e447d235a4d757a99f240ebbccd95922762b44008a", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cafc7ad3ca45b3e62147ad2d9609f9e706171c21c096884f7e6e1e6f4f64eeaa", "generation_seconds_batch": 21.255190233001485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3720580339431763, "target_z": 1.9765626213471268, "ce": 5.414999485015869, "genuineness": 4.570107460021973, "identity": 0.651661217212677, "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/emo_Confusion/997f7c17c63f755dfffde3e447d235a4d757a99f240ebbccd95922762b44008a.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "99b19490b1c99784832d86091fab73427f0493411525caa97a7d6bb2e2b3b4e2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4648633ef3200f00a89a9f97bab2e11561a43b4f912dc986102d2184488161dd", "generation_seconds_batch": 15.492544471053407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5845857858657837, "target_z": 2.3750001257121314, "ce": 5.808161735534668, "genuineness": 4.168440818786621, "identity": 0.7294411659240723, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/99b19490b1c99784832d86091fab73427f0493411525caa97a7d6bb2e2b3b4e2.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "df6bf9248b97d12571d44649b83cb236c9b1c05fd8d1df37b858abfaa053605c", "generation_seconds_batch": 15.407373536960222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06459558010101318, "target_z": -0.4746093900592657, "ce": 6.356205940246582, "genuineness": 4.557555198669434, "identity": 0.5551243424415588, "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/emo_Confusion/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "9aa3f5eb52cc2c8f44eeb313b6fdd33b661eff173ee424383a7dc5a28bf7cb96", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ba1d03617a5f8b402f68dd6de4e86a05caa0623c0261ed45bc1742b95c391c19", "generation_seconds_batch": 19.892604740918614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08959884941577911, "target_z": -0.4277343813292566, "ce": 6.205295085906982, "genuineness": 4.532451152801514, "identity": 0.3957088589668274, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9aa3f5eb52cc2c8f44eeb313b6fdd33b661eff173ee424383a7dc5a28bf7cb96.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "9b7a259b1b10804f872045a38d41a2a32ed30b9939f0e79d5dc1ef45f1cd3df0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2f77497ee7efb62ccd511386c14896de03060dc462ff7832d411c7ab2bc27c77", "generation_seconds_batch": 20.936262655071914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5000691413879395, "target_z": 0.3417969258523031, "ce": 5.974957466125488, "genuineness": 4.645420074462891, "identity": 0.7886166572570801, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9b7a259b1b10804f872045a38d41a2a32ed30b9939f0e79d5dc1ef45f1cd3df0.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "8e83b69c20c9c32b47d7412567bf77b6493bd9b2cf6a4669a657b8cc55b79849", "generation_seconds_batch": 15.53886979399249, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06042838096618652, "target_z": -0.48242186823424293, "ce": 6.252951145172119, "genuineness": 4.344170093536377, "identity": 0.6348050832748413, "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/emo_Confusion/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "9c1ba0e34fa45ce34d01cb49f180eedc73b1baac043d9328f9e2e2765744896a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f3a7caba0302fab1a1d6185523aa622eab6de7e2fa7e4a1f2c14845693ed3d27", "generation_seconds_batch": 19.343079576967284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09689147770404816, "target_z": -0.41406248865098816, "ce": 5.609594821929932, "genuineness": 4.457139015197754, "identity": 0.7527427673339844, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9c1ba0e34fa45ce34d01cb49f180eedc73b1baac043d9328f9e2e2765744896a.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "9c53d765d46a9446fb707c1b18e7ff5938cfe1b57d540e2888e62393f629d6b8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "314a8a1101bcd905da08b390f62d37316c600bd6efe0d7e0734edd77a552e154", "generation_seconds_batch": 20.83052072895225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16044144332408905, "target_z": -0.2949218612502356, "ce": 6.006728172302246, "genuineness": 4.570107460021973, "identity": 0.6506579518318176, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9c53d765d46a9446fb707c1b18e7ff5938cfe1b57d540e2888e62393f629d6b8.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "9c6b6afd581426a3f146ecb74f0e69f54350492b4a7e96f649d45562f9ea7a96", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b5db8723931fa5b93c29505f75680d6715bec1722d169e9216ba62b011e2b21e", "generation_seconds_batch": 21.00758527801372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16044144332408905, "target_z": -0.2949218612502356, "ce": 5.990842819213867, "genuineness": 4.758388519287109, "identity": 0.6454322934150696, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9c6b6afd581426a3f146ecb74f0e69f54350492b4a7e96f649d45562f9ea7a96.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "f2c5e3682e96dd91d02ce95996dc3ac3bf0f3270fce283dea6f9fb5a19f812ed", "generation_seconds_batch": 15.310805763001554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05105215311050415, "target_z": -0.5, "ce": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.6925469636917114, "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/emo_Confusion/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.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": "9d817ac0c4567afce72cbadd6415687a6c61e8c43d575eb8e12d15e5f0480dbe", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bbc60a8e65a517f5beac6c5a3973bd110e7eb36fed419bce4fe22147042cf0aa", "generation_seconds_batch": 14.405039991019294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2303727865219116, "target_z": 1.7109374694449682, "ce": 5.216432571411133, "genuineness": 4.482243061065674, "identity": 0.7294158935546875, "hypothesis": "Quelqu'un déplace la petite chaise bleue. Elle était ici près de la grande bibliothèque. Nous la remettrons après le déjà.", "wer": 0.14285714285714285, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9d817ac0c4567afce72cbadd6415687a6c61e8c43d575eb8e12d15e5f0480dbe.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "9dcfc6bfef2e68d0cd49e275652db18855acf99c8721a37b8a777ef7f24a019b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "88a8999c65c5597d2907699dd8f30fc30b87c3ee3cdc179e0a62c6e4653d9d18", "generation_seconds_batch": 21.863495272002183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2470417022705078, "target_z": 1.7421876056331103, "ce": 5.788304805755615, "genuineness": 4.469690799713135, "identity": 0.7917333841323853, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9dcfc6bfef2e68d0cd49e275652db18855acf99c8721a37b8a777ef7f24a019b.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "9e8cce80f1450a16dafef3be2eaa309b2930b965042119f96f18fedda86119eb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e89dbae0a9edd176c90653fad363c90a07812f99d01e9c3ddb06e8db5209c347", "generation_seconds_batch": 20.606669208966196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6803009510040283, "target_z": 0.6796874450009426, "ce": 5.816103935241699, "genuineness": 4.7081804275512695, "identity": 0.7736955285072327, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9e8cce80f1450a16dafef3be2eaa309b2930b965042119f96f18fedda86119eb.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "9e9bc786a68acb581fda3518f097e1e0d0d329ddf351613754eb680ef9cb0eb8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "83e11dfa814b43c38791a851c6a0804bcaa149645d443ab17a2920aa7bc7d29b", "generation_seconds_batch": 16.43362798599992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.357602596282959, "target_z": 0.07470703206843836, "ce": 5.414999485015869, "genuineness": 4.494794845581055, "identity": 0.5977764129638672, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wann 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/emo_Confusion/9e9bc786a68acb581fda3518f097e1e0d0d329ddf351613754eb680ef9cb0eb8.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": "9ef518c83ff630bee00115823520e7bf9fd68481bb787ba0db55121218327d8d", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "76bd457704a0cf02b21e0494dcfc71ddbedd901e05b3e4bc45c5041f3b92b01b", "generation_seconds_batch": 21.255190233001485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5323649644851685, "target_z": 0.4023436875804348, "ce": 5.871703147888184, "genuineness": 4.645420074462891, "identity": 0.8609341382980347, "hypothesis": "Quand quelqu'un a déplacé la petite chaise bleue, elle était ici, près de la grande bibliothèque, nous la remettrons après le déjeuner.", "wer": 0.047619047619047616, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9ef518c83ff630bee00115823520e7bf9fd68481bb787ba0db55121218327d8d.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "9fc28b964a3d9594f40443ba665f9dc7a05067e82b1e3a32ac592365cf7abbc1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6e2f1d3591dc757d3a2fa8c03b910b201f99b4929aebbca926d08a37583f9ccd", "generation_seconds_batch": 20.288248706026934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06042838096618652, "target_z": -0.48242186823424293, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.5508521795272827, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/9fc28b964a3d9594f40443ba665f9dc7a05067e82b1e3a32ac592365cf7abbc1.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "a076d22181ea81d589a4866cc2aaa21fce8f7dba9d806bca81a0cfa26cd67c50", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ab63dbeae9c2b15dc4c96fa29626b1421458a13cc954efdb661a6ffeb639b686", "generation_seconds_batch": 22.222612151992507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3995351493358612, "target_z": 0.15332030322436532, "ce": 5.5817952156066895, "genuineness": 4.595211505889893, "identity": 0.6701282262802124, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a076d22181ea81d589a4866cc2aaa21fce8f7dba9d806bca81a0cfa26cd67c50.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a08a24529049374300e20bcb20f406c347697066e0ad6ddfc453960310c18e13", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1eaaa3981ca2facec4af45982c4a03f4effcba18fad29bbfe49f658d10da6bf7", "generation_seconds_batch": 22.874549507978372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6803009510040283, "target_z": 0.6796874450009426, "ce": 5.673136234283447, "genuineness": 4.306513786315918, "identity": 0.8599046468734741, "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/emo_Confusion/a08a24529049374300e20bcb20f406c347697066e0ad6ddfc453960310c18e13.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": "a0c8bab752bb498e26be58e4b2724d0fb3b976d73c11545ca6d0797115578a42", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb692fea66738bd2f59f6992c3a2bfb91c2d9eb6e5c263f47cc14dd4981c5afc", "generation_seconds_batch": 16.207539557944983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3470547199249268, "target_z": 1.92968752880903, "ce": 5.903473377227783, "genuineness": 4.507347106933594, "identity": 0.6841868162155151, "hypothesis": "Mais ça j'est arrivé après mon départ je l'ai relu puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a0c8bab752bb498e26be58e4b2724d0fb3b976d73c11545ca6d0797115578a42.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a12cf88689b54004b927ab6276d55dcc5baf24adeec63f4b4cf5cd1c2ce8ee6a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "65a1e08e1b520e67adda1f29d9070e7ab2ad6669d2b25c7504cf4d98d56b3721", "generation_seconds_batch": 22.1693549240008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0428483486175537, "target_z": 1.35937505761806, "ce": 5.8875885009765625, "genuineness": 4.595211505889893, "identity": 0.6539284586906433, "hypothesis": "La puerta sigue abierta. Eh. 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/emo_Confusion/a12cf88689b54004b927ab6276d55dcc5baf24adeec63f4b4cf5cd1c2ce8ee6a.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a1709f0dbd034feb517985c560014ae3d0d3a354320e6fae99fa47820f74dfcf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "cb12b38ce6317a3cbca0f4e7bd3da81c1ca107e0b081750c3e26e2bf8ece2a5f", "generation_seconds_batch": 19.598603399004787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.209667205810547, "target_z": 3.5468747573057464, "ce": 5.3792572021484375, "genuineness": 4.287685871124268, "identity": 0.5897202491760254, "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/emo_Confusion/a1709f0dbd034feb517985c560014ae3d0d3a354320e6fae99fa47820f74dfcf.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "a1ca1d5fff76b0c53f0b583592f2b5e11c544da9c87f236a21c4d5f0662aef5f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d9183e95e30e9225bf833f753143029d916f13a9e083727b8a95361da8cddb1c", "generation_seconds_batch": 20.41972928098403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.229123115539551, "genuineness": 4.394378662109375, "identity": 0.4940509796142578, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a1ca1d5fff76b0c53f0b583592f2b5e11c544da9c87f236a21c4d5f0662aef5f.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a20d60bf12b6c5ce84d3148d17eb34d862b3b53e92d6b1fd632aaa286ae0e721", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "333cdd70209a36779953d25eda827eadc8ba85a0221b6ef0e56f4a4ce71ee5a2", "generation_seconds_batch": 19.598603399004787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5177797675132751, "target_z": 0.3750000139680146, "ce": 5.137005805969238, "genuineness": 4.532451152801514, "identity": 0.6150539517402649, "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/emo_Confusion/a20d60bf12b6c5ce84d3148d17eb34d862b3b53e92d6b1fd632aaa286ae0e721.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "a2453ccb2843619b01bd9490f547174a451125481351130f49fd76f1f108e4b4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a986175b4ce72fe084ee83fc71b36d11e2fd0fd605c7624d1f392a2f4839d43c", "generation_seconds_batch": 20.90115898195654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13543817400932312, "target_z": -0.3417968699802448, "ce": 5.967014789581299, "genuineness": 4.432034492492676, "identity": 0.7056849002838135, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a2453ccb2843619b01bd9490f547174a451125481351130f49fd76f1f108e4b4.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": "a26be300a2be173279d74fce77ca4f8ee67ca13070e3db515fdccdbab94515fc", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "004d56455f1c6fc5c06501ef08b2c4e6b0688cbc772de6b8a4f3b1fad205936e", "generation_seconds_batch": 21.81504573894199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3178842067718506, "target_z": 1.8749999580959562, "ce": 5.152891159057617, "genuineness": 4.406930446624756, "identity": 0.5252904295921326, "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/emo_Confusion/a26be300a2be173279d74fce77ca4f8ee67ca13070e3db515fdccdbab94515fc.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a284ac63471953918021804d38cdd4e6b4a90ca22be300df5eb09d0f7125d220", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2b50c9a529b1b6f2ad9eeef147147f3f37f2f9a0f66e274ecfe3e4c65a13a5d3", "generation_seconds_batch": 21.139076616964303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11356031894683838, "target_z": -0.3828124921429918, "ce": 6.173524379730225, "genuineness": 4.595211505889893, "identity": 0.6254334449768066, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a284ac63471953918021804d38cdd4e6b4a90ca22be300df5eb09d0f7125d220.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a3458ef76e471adfd6fc39e0221e81ea3eb75ca3d1455ff91694171506b0cc75", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd3e318db6568b989eaf9562e453bb872f9dfc13cda53677a3342fdf20bfc51a", "generation_seconds_batch": 14.91636687098071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5012415647506714, "target_z": 2.21875011523612, "ce": 5.526196479797363, "genuineness": 4.300237655639648, "identity": 0.6902384161949158, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a3458ef76e471adfd6fc39e0221e81ea3eb75ca3d1455ff91694171506b0cc75.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a371a93dd995de5eeb732945bfa0465ca62e254a76e1efe6cd4ea610d2f02566", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "402982ef32e43acb92132c111b1ba69e2b0073f0a318313c2ea99d4fbb64330c", "generation_seconds_batch": 15.608246298972517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2180018424987793, "target_z": 3.5625001606321676, "ce": 5.577824115753174, "genuineness": 4.482243061065674, "identity": 0.5695160627365112, "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/emo_Confusion/a371a93dd995de5eeb732945bfa0465ca62e254a76e1efe6cd4ea610d2f02566.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a39cf98f7439289017a51042fd2c20aa03b79083b1c83b90a5943080834a897f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ab9ccad849d177ab6aea1b30be124242818762194439a401328b9bebdf4e118e", "generation_seconds_batch": 21.37910819402896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4552716016769409, "target_z": 0.25781250611100637, "ce": 5.744620323181152, "genuineness": 4.570107460021973, "identity": 0.6710271835327148, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a39cf98f7439289017a51042fd2c20aa03b79083b1c83b90a5943080834a897f.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "a3de96df7eadba68ad38d6369bf701051aa52b2eb01423ad48d69933f385f342", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d6d77ccdc0627b24d3816d096643e7e934541fa579e3f4055c4010715c6213e6", "generation_seconds_batch": 20.936262655071914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7428091764450073, "target_z": 0.7968750646020675, "ce": 5.689021587371826, "genuineness": 4.68307638168335, "identity": 0.7083637714385986, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a3de96df7eadba68ad38d6369bf701051aa52b2eb01423ad48d69933f385f342.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "a41ed9888a6f96287ed0ae743aec2a73fdfada98edcefb78bbe5a4502a788ada", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e57fc3043b2ad29c798a8d54988ee4d1d021b5356afc1f983f5dc8ae67ff2c35", "generation_seconds_batch": 19.616920884000137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.25159916281700134, "target_z": -0.12402345419614244, "ce": 5.593709468841553, "genuineness": 4.7332844734191895, "identity": 0.7980282306671143, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a41ed9888a6f96287ed0ae743aec2a73fdfada98edcefb78bbe5a4502a788ada.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "a4789b623cd295676a370f9b686b942798d198ebe7b9b84250757aa38c6f8bb6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f9bf30f4607bc45f718683d1a6e1efebde48a92d51754ba9f7ab2d6e625e32ad", "generation_seconds_batch": 21.75843370903749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6345922946929932, "target_z": 2.468750087300091, "ce": 5.831989288330078, "genuineness": 4.482243061065674, "identity": 0.5698181390762329, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a4789b623cd295676a370f9b686b942798d198ebe7b9b84250757aa38c6f8bb6.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a486143d3fc863c3be65b11a7a84419d9d060799e452b4407567612e54773d24", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f02606a3319cac88abed2a6c77cf3d643aa8d96ac6e9fc5e169e1bd72819a05a", "generation_seconds_batch": 20.652124678948894, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4513654708862305, "target_z": 3.999999832383825, "ce": 4.894754409790039, "genuineness": 4.231201171875, "identity": 0.542057454586029, "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/emo_Confusion/a486143d3fc863c3be65b11a7a84419d9d060799e452b4407567612e54773d24.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "a48b04b561040fd636ebfe46bb034e6612ffd98da844e8c8d7d4c155b76bded6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "50417450911047d1d80d020bbb4f0fc6b05376b8ec2ff64f22d93cb1814b522b", "generation_seconds_batch": 15.923169014975429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1677340567111969, "target_z": -0.28124999650799637, "ce": 5.967014789581299, "genuineness": 4.256305694580078, "identity": 0.24144433438777924, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a48b04b561040fd636ebfe46bb034e6612ffd98da844e8c8d7d4c155b76bded6.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "a49c79b357ca58b8c7a9e9a00a75299e57c5ed06f1219d905e6e4b80ea4ffdf3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d04c785b0d246c6c743dc8e2bc29b00316fa3cf2cd5757fe1abd763e6253c32c", "generation_seconds_batch": 19.668697881978005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3115028142929077, "target_z": -0.011718773134524161, "ce": 6.062326908111572, "genuineness": 4.3127899169921875, "identity": 0.3396543562412262, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a49c79b357ca58b8c7a9e9a00a75299e57c5ed06f1219d905e6e4b80ea4ffdf3.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "a4ca08c80f86a62e5c8e47082bd983e41c647ba845291fa6f0962b52d001b3fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b02942e501646525b24eee628f24faecd16780ad1c663a26608d08df2bb6c04b", "generation_seconds_batch": 20.38351629592944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 5.927301406860352, "genuineness": 4.758388519287109, "identity": 0.7012218832969666, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a4ca08c80f86a62e5c8e47082bd983e41c647ba845291fa6f0962b52d001b3fe.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a4e6038bb3ceae83f3a720dea206dc7683a9f5d3d017af8da1f0a1fc111aa346", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "60f50182c111f54930a37fb38dfddfe3b76ea8b41598f68846fabd8d6c885ad7", "generation_seconds_batch": 24.067310199025087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 5.935244083404541, "genuineness": 4.319066047668457, "identity": 0.6310697793960571, "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/emo_Confusion/a4e6038bb3ceae83f3a720dea206dc7683a9f5d3d017af8da1f0a1fc111aa346.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "a52d6f5de1fb66ae14f4d16c1b78c6c27e6b2825c042ac131acb92ae55eb384c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "07347a32dc6c3f3544f13c7173eed3c0f72e47ba78a5a88d9344b4b361ce1a72", "generation_seconds_batch": 20.532968975952826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6013851165771484, "target_z": 4.281249940635938, "ce": 5.198561668395996, "genuineness": 4.432034492492676, "identity": 0.6269181966781616, "hypothesis": "La puerta sigue 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/emo_Confusion/a52d6f5de1fb66ae14f4d16c1b78c6c27e6b2825c042ac131acb92ae55eb384c.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "a5318488dbc28a313b4e76601e5a282fa8ff9942dcb681928783d78ceea28d7b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "583d6144523365a55791215521402b5dbacbafd4b6de10e3c7b5422613cf861a", "generation_seconds_batch": 22.14455138600897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9428352117538452, "target_z": 1.171874910953907, "ce": 5.25614595413208, "genuineness": 4.394378662109375, "identity": 0.7252311110496521, "hypothesis": "J'ai trouvé la petite boîte. Elle était près 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/emo_Confusion/a5318488dbc28a313b4e76601e5a282fa8ff9942dcb681928783d78ceea28d7b.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a562de826e8588b869d493c35519f244b3919ae10641578d0f6ad6c6aac8e8be", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bb8551fd1071dc4ccb179da082b78513b0eff99bb70ec4a890849ce0b57a85f7", "generation_seconds_batch": 15.826243589050137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4302683472633362, "target_z": 0.21093752531702645, "ce": 5.720792293548584, "genuineness": 4.65797233581543, "identity": 0.5165744423866272, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a562de826e8588b869d493c35519f244b3919ae10641578d0f6ad6c6aac8e8be.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "a56b42d4583317b9d6838ef134a36bd1753025a6e15099f91b259fe666f5f31a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f6d96d2cfa37fb749a08b353f6a00f30c8f18b13389f6c732a2cf14677677617", "generation_seconds_batch": 21.136447174008936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.011456012725830078, "target_z": -0.6171875078570082, "ce": 6.062326908111572, "genuineness": 4.545003414154053, "identity": 0.7589248418807983, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a56b42d4583317b9d6838ef134a36bd1753025a6e15099f91b259fe666f5f31a.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "a56c7ab399490ac70a7e11e8dd8872793786aedbc402643409316ee4ee228fd9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31f20c9e38f5e7c6bf3faf61bbfd0a51d4a731eadae3e96f3d0699b66c9b887c", "generation_seconds_batch": 16.372532495995983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6990534067153931, "target_z": 0.7148437085324567, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.5952061414718628, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a56c7ab399490ac70a7e11e8dd8872793786aedbc402643409316ee4ee228fd9.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "a5a6428ef9390826b6a269a02da4052e7786e1fe4006bee2c382d12d7af79e90", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5020a68b0026b4482b772d00966a122ce5a0b5e6240a326b98522f550a3ae80a", "generation_seconds_batch": 21.811032108031213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16044144332408905, "target_z": -0.2949218612502356, "ce": 5.998785495758057, "genuineness": 4.833701133728027, "identity": 0.606523871421814, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a5a6428ef9390826b6a269a02da4052e7786e1fe4006bee2c382d12d7af79e90.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "a5c0cb5a81169cafde62c4cfb56d9502e1ef7bf53f6c681816e20bc92ae6a897", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6ad8217bdaa73b442b342282edf211267b721b9bb0a81142149bf254e3aea648", "generation_seconds_batch": 15.389283393044025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9261664152145386, "target_z": 1.1406249982539982, "ce": 5.8875885009765625, "genuineness": 4.218649387359619, "identity": 0.6741495728492737, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a5c0cb5a81169cafde62c4cfb56d9502e1ef7bf53f6c681816e20bc92ae6a897.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "a67d8fe4c47b96e96204ea88d44f53a76fda874d173907761492c26928a6296e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f1996384a8ddcf269abb7090de323895e7dc32908b4e1c3033e6e88b0505da85", "generation_seconds_batch": 19.111097343964502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5345792770385742, "target_z": 2.2812501641241716, "ce": 5.895530700683594, "genuineness": 4.281409740447998, "identity": 0.4988762140274048, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a67d8fe4c47b96e96204ea88d44f53a76fda874d173907761492c26928a6296e.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "a6a1bc23fc8d7adc8fac45c7cd7b3f3e048c5154cafe22c906e9413160320adc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "747c175bf166dcef3971bbcb3b5036847d9f668473feb44082366e0b9cd769e9", "generation_seconds_batch": 21.999544403981417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1187693327665329, "target_z": -0.3730468664882411, "ce": 6.109982967376709, "genuineness": 4.557555198669434, "identity": 0.7721114158630371, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a6a1bc23fc8d7adc8fac45c7cd7b3f3e048c5154cafe22c906e9413160320adc.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "a775d2018039d4d20de5bd6d21e7117603f4637cdcd01209707f48fd983ee411", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e2b1e60ef44ee95f7cabcc2c02ecba370a31efb47e30fdde8072b1dfe26ce348", "generation_seconds_batch": 22.51326278306078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.988674521446228, "target_z": 1.2578123943668897, "ce": 6.308549880981445, "genuineness": 4.6956281661987305, "identity": 0.7784606218338013, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a775d2018039d4d20de5bd6d21e7117603f4637cdcd01209707f48fd983ee411.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "a78319c609090944296529dcfaab9f2272ad40adf74eb20440d9a98c256ab556", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2e710a3406c9994412e1aba6cf9b97be3bba10e67ee898c2c20b3d3a47a007ef", "generation_seconds_batch": 25.8729525039671, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5584100484848022, "target_z": 0.4511718437902174, "ce": 5.780362129211426, "genuineness": 4.6705241203308105, "identity": 0.7481215000152588, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a78319c609090944296529dcfaab9f2272ad40adf74eb20440d9a98c256ab556.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a7dee3be20379612546dd433a69395c109d93164cd35c4cfd480e5d83cf8a163", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a903ed7a38e29c9badb4dd7927bfb8915ace51e37a32de350a0482f2e1e50246", "generation_seconds_batch": 21.73185832996387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34015241265296936, "target_z": 0.04199220987064836, "ce": 5.80021858215332, "genuineness": 4.457139015197754, "identity": 0.5207530856132507, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a7dee3be20379612546dd433a69395c109d93164cd35c4cfd480e5d83cf8a163.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "2bada9bc753409fe2820b8037ba3a2b762a5aa767185462a9982767761944372", "generation_seconds_batch": 15.188428443972953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021881669759750366, "target_z": -0.5546875148410155, "ce": 6.133810997009277, "genuineness": 3.9299514293670654, "identity": 0.7248326539993286, "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/emo_Confusion/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "0663ce16de6270c60e1217a8fcaf3c137ba4123cfb404cf3a227b70f03eb094b", "generation_seconds_batch": 15.293119743932039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.013547241687774658, "target_z": -0.5703125270630283, "ce": 6.157639026641846, "genuineness": 4.607763767242432, "identity": 0.6374467611312866, "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/emo_Confusion/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "a95829e50e73108934c8e969cb2c22363d8ea5e17dd999d8392c96ed56affa95", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a1b08dfa1f95f5ae550d059b6f2c4e057cef954040d7c825baebb056f285425b", "generation_seconds_batch": 20.82704621704761, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8553237915039062, "target_z": 1.0078124223029188, "ce": 5.7684478759765625, "genuineness": 4.400654315948486, "identity": 0.6432791352272034, "hypothesis": "Die Tür 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/emo_Confusion/a95829e50e73108934c8e969cb2c22363d8ea5e17dd999d8392c96ed56affa95.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "a95a215d61eab7e96642333b65c2b2360e73125b6742dc4e35860f9c5a033a6d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18c997bc4562a6c032dde82bb6c0c9b56c76cd27e52b2f07fce3d98f38b1d4b6", "generation_seconds_batch": 22.11482066300232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5073617696762085, "target_z": 0.35546881853057155, "ce": 5.692992687225342, "genuineness": 4.595211505889893, "identity": 0.6388117074966431, "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/emo_Confusion/a95a215d61eab7e96642333b65c2b2360e73125b6742dc4e35860f9c5a033a6d.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "a97bf4e3184a5fb72e815f558447bbd7e557523399fdd075cb7a724d62d747a6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bcec90a27cb6948f24e93655861975a04129b4ae612fd8637d139b33ff38a6c1", "generation_seconds_batch": 16.48558146005962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1263229846954346, "target_z": 3.3906247468297357, "ce": 5.720792293548584, "genuineness": 4.086852550506592, "identity": 0.38572412729263306, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a97bf4e3184a5fb72e815f558447bbd7e557523399fdd075cb7a724d62d747a6.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "a9b538d3cf40ddd64daaed257bc4555f665847e2eceb3538ed568b46e653a4e6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e7cd48696cca56b93265c2742a2e4111327124fcd5b585d69b9585dccb666c52", "generation_seconds_batch": 15.659956371993758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5262447595596313, "target_z": 2.2656249842859837, "ce": 5.561938762664795, "genuineness": 4.519899368286133, "identity": 0.7576463222503662, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/a9b538d3cf40ddd64daaed257bc4555f665847e2eceb3538ed568b46e653a4e6.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "aa0e9cadfdc9b8f6ac53b1f3bb37b10625e4e5f2814fe859ce206a9ed336dcd6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1ac00494e6c21ffd77f87b3fe25544325cdd8e29aeb607ef96317078368be5ed", "generation_seconds_batch": 20.419816776993684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 5.982900142669678, "genuineness": 4.758388519287109, "identity": 0.7365965843200684, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/aa0e9cadfdc9b8f6ac53b1f3bb37b10625e4e5f2814fe859ce206a9ed336dcd6.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "aafa279e70e7955cddc9da3061156ac64016b2a1aa2f8b1f32894c44387e406e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f3218d484ac4fdc499a33116dfe70991aee33512d9a36e2b5f59b38898a82d20", "generation_seconds_batch": 20.353285230929032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9303336143493652, "target_z": 1.1484374764289753, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.4340458810329437, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/aafa279e70e7955cddc9da3061156ac64016b2a1aa2f8b1f32894c44387e406e.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "ab0571ecf50d9ccdd3f57e152b49224e238fa2f6724b584e338e225becce31c6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc62a78301903224f75e4c832900af1e5c43022dde63cf702fc9f9ecac96340e", "generation_seconds_batch": 14.786553442943841, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6511304974555969, "target_z": 0.6249999860319854, "ce": 5.414999485015869, "genuineness": 4.645420074462891, "identity": 0.695993185043335, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la vida.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ab0571ecf50d9ccdd3f57e152b49224e238fa2f6724b584e338e225becce31c6.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ec564cfbfc9fe2f1e8cf1514ff9b6347b5e40ec908426a289e3e72df92c470dd", "generation_seconds_batch": 15.203101939987391, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0698046088218689, "target_z": -0.46484373646848587, "ce": 6.427689552307129, "genuineness": 4.21237325668335, "identity": 0.3091121315956116, "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/emo_Confusion/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "aba3f483f79cca84fbc2f703edce9a3516aa676e5d52c42cf821266364cea959", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f4fc81e533f3f70e1d7d908ae61212ec7d7cea0fa47e1e0b07e0f897a8c544c0", "generation_seconds_batch": 15.820332654053345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9679691791534424, "target_z": 3.093750129204135, "ce": 5.629451274871826, "genuineness": 4.20609712600708, "identity": 0.3610304594039917, "hypothesis": "Er staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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/emo_Confusion/aba3f483f79cca84fbc2f703edce9a3516aa676e5d52c42cf821266364cea959.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": "abc7ee353798a9d77ad1f63370d1a95182a8aff1dba464251e4c32015ff5fa10", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f04d5db8fa5ae0473b9f0e08fa26806dd8370fedceeef665857d9a89bfbf7b9", "generation_seconds_batch": 21.81504573894199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9470024108886719, "target_z": 1.1796873891288842, "ce": 5.7128496170043945, "genuineness": 4.381826400756836, "identity": 0.6956371068954468, "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/emo_Confusion/abc7ee353798a9d77ad1f63370d1a95182a8aff1dba464251e4c32015ff5fa10.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "acc8f14273bd314ffea6cadeced2dfb1d521aeb868aaf4877f6d921e44bd4881", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d0ef4ca5ab887b2c54956bb22b64ec4698f318356debddc67d583c8f37ff98b", "generation_seconds_batch": 19.042343620909378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7429397106170654, "target_z": 2.6718749668259654, "ce": 5.363371849060059, "genuineness": 4.545003414154053, "identity": 0.6405602097511292, "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/emo_Confusion/acc8f14273bd314ffea6cadeced2dfb1d521aeb868aaf4877f6d921e44bd4881.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "ad6dd3d3d266ca4f699d9b1f3cd879525e6c5ddbbd7af01f32f6d3ce67d39201", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3218832018a3695a513b1e536b28282b52c4406624286c6ed6d0d95e0a61383d", "generation_seconds_batch": 21.079515608027577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.296917587518692, "target_z": -0.03906250261900274, "ce": 6.09409761428833, "genuineness": 4.570107460021973, "identity": 0.47773706912994385, "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/emo_Confusion/ad6dd3d3d266ca4f699d9b1f3cd879525e6c5ddbbd7af01f32f6d3ce67d39201.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "ad9fa6e1e91146ae865e46c6bb0c98a795dffe4c9d9a13c70237dc2f669e5029", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "abfa2fcf43a8a5bbae5a82e30975e8a2ea7fa7b6568c104fe3d1ecd4680c9e80", "generation_seconds_batch": 22.114413153962232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4042232632637024, "target_z": 0.16210936910724386, "ce": 5.95112943649292, "genuineness": 4.419482707977295, "identity": 0.5845543146133423, "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/emo_Confusion/ad9fa6e1e91146ae865e46c6bb0c98a795dffe4c9d9a13c70237dc2f669e5029.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "ada501355aeef83a407f7f9ed60b4ab7e0354e26f08f7c26e831686b16fc503b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "758cd1212c41756c4212626f169eec09a45deb2ad2c7dcc288f77a47a60bacec", "generation_seconds_batch": 21.017572115990333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261794328689575, "target_z": 1.328124921429918, "ce": 6.205295085906982, "genuineness": 4.375550270080566, "identity": 0.4271288812160492, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ada501355aeef83a407f7f9ed60b4ab7e0354e26f08f7c26e831686b16fc503b.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "11a82ee628948df57a6e8771f45421adbc6f813d64808ab8d2a1ff61590d13f4", "generation_seconds_batch": 15.878778078011237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2083643674850464, "target_z": -0.2050781108137352, "ce": 6.181467056274414, "genuineness": 4.381826400756836, "identity": 0.7055938243865967, "hypothesis": "I bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar, wanneer we kunnen praten?", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "551f45d7e9b8e3d852d2be5eb1af21d2c5e72db281aeadecd3ea7f9dbedbee9a", "generation_seconds_batch": 16.154919733991846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 6.483288764953613, "genuineness": 4.099404335021973, "identity": 0.1711663454771042, "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/emo_Confusion/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.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": "af612b424088b93607a43110ada3150dd9d6112eb6619973559f2b01f5aa8e10", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b468cda3ea4a4045a3c76248687e21e27884df065530bbda8dc6dd327387c038", "generation_seconds_batch": 16.487892873003148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8449058532714844, "target_z": 0.9882813386095926, "ce": 5.772419452667236, "genuineness": 4.231201171875, "identity": 0.4263530969619751, "hypothesis": "I wrote the number on the envelope. Write it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/af612b424088b93607a43110ada3150dd9d6112eb6619973559f2b01f5aa8e10.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "afe660b46ca741ba86e71460b07a1e242ff66df3809337f5637b263f686ce91f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f4498808686c797f25a8fed885fce487b1f188dbeae907443d07774f48dcb636", "generation_seconds_batch": 22.386052491958253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.592920184135437, "target_z": 2.3906250820620856, "ce": 5.927301406860352, "genuineness": 4.356721878051758, "identity": 0.4660831689834595, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/afe660b46ca741ba86e71460b07a1e242ff66df3809337f5637b263f686ce91f.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "b094b98af805f4292ad62e89061d8736ab5e9bfccc15c50bf767ad66973f9979", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5f0fcfb83088c02f910c322ee69f31979e97fc7f0cf28e12381af0309541beb9", "generation_seconds_batch": 22.45731294795405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 5.982900142669678, "genuineness": 4.7081804275512695, "identity": 0.6973048448562622, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b094b98af805f4292ad62e89061d8736ab5e9bfccc15c50bf767ad66973f9979.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": "b10dec8e6d48a2b57b172251302b2e171a2e6f097a21823baf473e292ec81266", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "199cf444e8bb316b36b65d9306f6d322b4cd17259930f9e4c310464b89081f9d", "generation_seconds_batch": 22.034487597993575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2513394355773926, "target_z": 3.6249999860319853, "ce": 5.40705680847168, "genuineness": 4.432034492492676, "identity": 0.6009276509284973, "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/emo_Confusion/b10dec8e6d48a2b57b172251302b2e171a2e6f097a21823baf473e292ec81266.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "b130f7d0567233c656d5bbb583cb1de79bbe943109eee56dedcf98840f11a872", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1d4671a80691363505ae096535134a40c7e928a55a26a4ac6307dca04c82a218", "generation_seconds_batch": 16.38850259105675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846118211746216, "target_z": 2.749999972063971, "ce": 5.458683967590332, "genuineness": 4.18726921081543, "identity": 0.6153336763381958, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b130f7d0567233c656d5bbb583cb1de79bbe943109eee56dedcf98840f11a872.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "b2926d00a0c8228c76a54304e9595019a32ce7c58489a5f620525628e989ee37", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c3eae9396d36bca9ffc185e1555fb2b4c096f8946ede288ede7b0cd55f3abc2d", "generation_seconds_batch": 16.372532495995983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.451235055923462, "target_z": 2.1250001536481604, "ce": 5.80021858215332, "genuineness": 4.419482707977295, "identity": 0.27916383743286133, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b2926d00a0c8228c76a54304e9595019a32ce7c58489a5f620525628e989ee37.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "b318ab71f5d47c1714e38f73a2c6db570b602e80826cddc7a6dd2617a6c5b02c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "840e18bc9486df3b1fba9980371cf27d3d3f088c3a42aee8faa194490c5e7b46", "generation_seconds_batch": 20.042186809005216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0761860609054565, "target_z": 1.4218751065061113, "ce": 6.0305562019348145, "genuineness": 4.519899368286133, "identity": 0.5282384157180786, "hypothesis": "I found the l 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/emo_Confusion/b318ab71f5d47c1714e38f73a2c6db570b602e80826cddc7a6dd2617a6c5b02c.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "b32dba2de35b6193653e3d7de4fe2110ac09f9c8bb7e871b119bdf489e016976", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e82be28773d83350935db9ce197ae7f58b909e8ddfa4d284b55c9cfd0ca29267", "generation_seconds_batch": 25.187727207085118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7596086263656616, "target_z": 2.7031251030141075, "ce": 5.593709468841553, "genuineness": 4.168440818786621, "identity": 0.5393615961074829, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Nienmantel 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/emo_Confusion/b32dba2de35b6193653e3d7de4fe2110ac09f9c8bb7e871b119bdf489e016976.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "b35b792107950be029a4a9e93e537082e4f54abb61bdf59e42cb4a0a8a2290bb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f29bfa71c39da4fb3d0cce49fbb9b3c953820b515c97019aa1e16c5fc55b5988", "generation_seconds_batch": 14.85682431096211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0345139503479004, "target_z": 1.3437501012681057, "ce": 6.09409761428833, "genuineness": 4.507347106933594, "identity": 0.6487593650817871, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b35b792107950be029a4a9e93e537082e4f54abb61bdf59e42cb4a0a8a2290bb.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "b375c1d9c75ccd15133c80920a30f74ab365c9f1a6427e6043bccabf029dcc7b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "fa293fd18219657df0456ba3b76bf67490561d1b4bd7499b16e1aaa32bc0e236", "generation_seconds_batch": 19.690413024043664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.376225233078003, "target_z": 1.984375099522104, "ce": 6.1020402908325195, "genuineness": 4.432034492492676, "identity": 0.6437699794769287, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b375c1d9c75ccd15133c80920a30f74ab365c9f1a6427e6043bccabf029dcc7b.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "1ffcd317d00b2f0874f23b8bb007ce084cf7302cc71904d1f894136066d8c29b", "generation_seconds_batch": 16.30192103504669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.316492557525635, "genuineness": 4.105680465698242, "identity": 0.5400815010070801, "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/emo_Confusion/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "b4abd02c98b4333bb040f8d7644c56f2179cad6990a82d99aaac244f18bd6a58", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a4267413144b88bcecc844c5ba3ce88a3f872ae1005747e6a49959a5d24d4b41", "generation_seconds_batch": 20.135928080999292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 5.8875885009765625, "genuineness": 4.557555198669434, "identity": 0.7550026178359985, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b4abd02c98b4333bb040f8d7644c56f2179cad6990a82d99aaac244f18bd6a58.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "b5582d52e05e213b9294946a9063544fbbcfb4846fb946af31151c42efe890c5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d4179ae239e0b880c1c220b7ac920a477095f86061bf9adcccc1aa58c7c06f3b", "generation_seconds_batch": 21.75843370903749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.88462495803833, "target_z": 2.9375001187281238, "ce": 5.732706069946289, "genuineness": 4.055472373962402, "identity": 0.12107210606336594, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b5582d52e05e213b9294946a9063544fbbcfb4846fb946af31151c42efe890c5.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "b57a57b2a9f331fb3bb4865e8df0490ad90da1546666b66a3c47135b2f02a2c0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "de358acf5d04cfb19f53c153d5b6773620f1fb6bc2f37bbf2aef32045b9cfd7e", "generation_seconds_batch": 19.44128825608641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0553499460220337, "target_z": 1.3828124921429918, "ce": 5.95112943649292, "genuineness": 4.7332844734191895, "identity": 0.7854040861129761, "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/emo_Confusion/b57a57b2a9f331fb3bb4865e8df0490ad90da1546666b66a3c47135b2f02a2c0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "b59c7b17d1417327eb0b7d9258431422aea2469a5b5d3f7c447f0178e4fa9a35", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c658785a7ebc35728f6f17e66dd1acd681c06a0df48670c671716adfa0400fb0", "generation_seconds_batch": 21.024782811990008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2620171904563904, "target_z": -0.10449220288664107, "ce": 5.80021858215332, "genuineness": 4.6705241203308105, "identity": 0.6897361278533936, "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/emo_Confusion/b59c7b17d1417327eb0b7d9258431422aea2469a5b5d3f7c447f0178e4fa9a35.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "b59fd7563efaa3da38b43df7d996936e151168dc0dfbcb81a10834ca51bd8538", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b12e038b23fcabdb8b332b1857e0dc5ea4476a444c651b3e27168a92911df859", "generation_seconds_batch": 21.57632663694676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2720448970794678, "target_z": 1.7890624746829735, "ce": 5.478540420532227, "genuineness": 4.457139015197754, "identity": 0.6165226101875305, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b59fd7563efaa3da38b43df7d996936e151168dc0dfbcb81a10834ca51bd8538.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "b5be13640cc5afdcdfba6806af922b2f629b25c4aa2babe750dfc8597a19a927", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7ea45a8541043a2036cffebd5fe4f8cf2f6c79efc6a1eeb6f306bc17c0df4a65", "generation_seconds_batch": 15.989142192993313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4089113771915436, "target_z": 0.1708984349901224, "ce": 6.300607204437256, "genuineness": 4.6203155517578125, "identity": 0.32262545824050903, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b5be13640cc5afdcdfba6806af922b2f629b25c4aa2babe750dfc8597a19a927.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "b5d546a648f7ae2e04f0fddcfb219d26d02d7a712bfe2ce4a9e4c0a81ea1f6dc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "69f719bfe11f6e444f13d020a266c50eca6e5d76a873e7cd32331bc7fb0a3a90", "generation_seconds_batch": 15.718381457962096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.517910361289978, "target_z": 2.250000027936029, "ce": 6.006728172302246, "genuineness": 4.432034492492676, "identity": 0.7297797203063965, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b5d546a648f7ae2e04f0fddcfb219d26d02d7a712bfe2ce4a9e4c0a81ea1f6dc.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": "b5dfa48a1c1f82d3307e9cee4795c68a120bfb52e9f8574c8ecc7e03d93adfd3", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ccd6a42b76d1ff4b9887d4f8a441935694a9b1d5e944a07f7adeb95f67a741b", "generation_seconds_batch": 16.275952197960578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2303727865219116, "target_z": 1.7109374694449682, "ce": 5.276002407073975, "genuineness": 4.331617832183838, "identity": 0.5666936635971069, "hypothesis": "La voûte tourne 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/emo_Confusion/b5dfa48a1c1f82d3307e9cee4795c68a120bfb52e9f8574c8ecc7e03d93adfd3.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "b66c3d69e66157a463ae49df03aa6aaba63e5f72c60788f901600ba95a2e0f21", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2cfd46ade314453df00c14a8b22eb753996d181de8fc67e0d7df3e4860f481df", "generation_seconds_batch": 14.834789806045592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3289530277252197, "target_z": 0.02099610493532418, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.5302520394325256, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b66c3d69e66157a463ae49df03aa6aaba63e5f72c60788f901600ba95a2e0f21.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "b6a19d4e182d586f3ce28b32ad006c5753d6d4d3f5e98ede504375e688e9251d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a7a8a0a6f10857af8a4bb2bdc55f3c8fdc28438f080f7c333dd2b88d31a14c1", "generation_seconds_batch": 21.143578093964607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.48027485609054565, "target_z": 0.3046874869049863, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.24063768982887268, "hypothesis": "the door is still open. I put your coat on the chair. You tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b6a19d4e182d586f3ce28b32ad006c5753d6d4d3f5e98ede504375e688e9251d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "b71c111368e66d1d68b45f08d1b65c5f445863bd66bc3f2bb701664067288b27", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "dbe5e2bcbd3fb549b8ca442da9e73728d9e7fbaa7c335944eb0a9b0ac944298c", "generation_seconds_batch": 22.169447844964452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.011463642120361328, "target_z": -0.5742187661505168, "ce": 6.022613525390625, "genuineness": 4.570107460021973, "identity": 0.595197856426239, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b71c111368e66d1d68b45f08d1b65c5f445863bd66bc3f2bb701664067288b27.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "b7e9691242bbf45d0c3af971b2fe58e03534abbf8519b8fee3b7e946582f9c0c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f4dbe654141284020604f167383d1f3f74b71f1c16fe58d789a67e6ad3666543", "generation_seconds_batch": 19.256980390055105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39015892148017883, "target_z": 0.13574217145860826, "ce": 5.760505676269531, "genuineness": 4.400654315948486, "identity": 0.6550761461257935, "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/emo_Confusion/b7e9691242bbf45d0c3af971b2fe58e03534abbf8519b8fee3b7e946582f9c0c.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "b83470fe46797528891962ad2ec4b2f3bfbff7a296edd47f9db0132526451961", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "440ef0b43999944a1d7765887e3c3baf6f9a2ac6adfbbfb8c42875cbe8eb9544", "generation_seconds_batch": 21.553457865025848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8762905597686768, "target_z": 2.9218751623781696, "ce": 5.295859336853027, "genuineness": 4.231201171875, "identity": 0.5495626330375671, "hypothesis": "De de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b83470fe46797528891962ad2ec4b2f3bfbff7a296edd47f9db0132526451961.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "b97045d5e702b6321621493741e58e308b68a55cbadaa4f3f36bb0311f35b838", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7b3725bb2144ab2a03124908a0f0d13a4e7fab7d5b940044497852df73fbc038", "generation_seconds_batch": 19.343079576967284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8469893932342529, "target_z": 0.9921874659529645, "ce": 5.998785495758057, "genuineness": 4.758388519287109, "identity": 0.8019943833351135, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/b97045d5e702b6321621493741e58e308b68a55cbadaa4f3f36bb0311f35b838.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "a7c8fe090262234577032897355bcff2e4285500caaed2dd02b2363025a15e95", "generation_seconds_batch": 15.621403019991703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04063412547111511, "target_z": -0.5195312513095014, "ce": 6.117925643920898, "genuineness": 4.444586753845215, "identity": 0.7872172594070435, "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/emo_Confusion/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "bab94321097e8fd00fdfdf259b855ff0ab8360d434fae6dde6f6e55abb1ba1d4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7772b148c0efad1feacfdc1f0d5ef53c3c2d288d1408809171ba7e15da42830d", "generation_seconds_batch": 23.698289455031045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18857011198997498, "target_z": -0.2421874938889936, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.742730975151062, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bab94321097e8fd00fdfdf259b855ff0ab8360d434fae6dde6f6e55abb1ba1d4.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "bacec14c3d892ce2a0b9583ad7ce69c9d90a3d0eb29d05ddb0ec7f74b2870a3b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fff6d2e868ea74a032251625122510af16c034475998f6442ce229bc45ef189d", "generation_seconds_batch": 20.95043559395708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6052911877632141, "target_z": 0.5390625026190027, "ce": 6.173524379730225, "genuineness": 4.425758361816406, "identity": 0.3263857960700989, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bacec14c3d892ce2a0b9583ad7ce69c9d90a3d0eb29d05ddb0ec7f74b2870a3b.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "bb02d0ada7e4a7832d366f366dbc40b53bae6321b101399074e947c54abac121", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "77fa484bcbff1c794878239a398b1a72fcd4f25fbe6b6bc9e571868a162a585d", "generation_seconds_batch": 21.294460720033385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 6.070269584655762, "genuineness": 4.7834930419921875, "identity": 0.742490291595459, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bb02d0ada7e4a7832d366f366dbc40b53bae6321b101399074e947c54abac121.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "bb39c9900438608f06a11bc361fcb78effbf4b345cbdb9c93d4e5dfde6969022", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "fbe12c924cf39b48b107d9c17a3c80e2e237e59325f88d450e8edbe4b0587db8", "generation_seconds_batch": 14.863663148018532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1937791258096695, "target_z": -0.23242186823424293, "ce": 6.022613525390625, "genuineness": 4.607763767242432, "identity": 0.6062914133071899, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bb39c9900438608f06a11bc361fcb78effbf4b345cbdb9c93d4e5dfde6969022.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "bc0d63a22081c79e1f742e855f9167f3d6c44256bb9a719805394bbf65d96d22", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "177348d49c4188ea8ff5d9566420c5c491ae812725257e52ed130b98eebfc6b7", "generation_seconds_batch": 17.154648658935912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06147018074989319, "target_z": -0.48046874869049866, "ce": 5.990842819213867, "genuineness": 4.6956281661987305, "identity": 0.754653811454773, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bc0d63a22081c79e1f742e855f9167f3d6c44256bb9a719805394bbf65d96d22.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "bc220dbdf195b365b78fa252a80de18b02c6afb0231957ca4b7f2032c6500b06", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "41f771b9092350b51703cb453e0166d67b0c4b52eb7a84fe947b438e9793bdc4", "generation_seconds_batch": 14.70974964601919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4845726490020752, "target_z": 2.187499979047978, "ce": 6.078212261199951, "genuineness": 4.137060642242432, "identity": 0.6231157183647156, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bc220dbdf195b365b78fa252a80de18b02c6afb0231957ca4b7f2032c6500b06.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "bca6c4d4d804fef0bb6b8b5cb78cd2d965771a6493fd2d160f41370312166a59", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "433c3ecad31769be940ac8cc32b07f27e3a858310f1e604e2678117fe98f7ca4", "generation_seconds_batch": 13.980686991009861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04271772503852844, "target_z": -0.5156250122220127, "ce": 6.117925643920898, "genuineness": 4.7332844734191895, "identity": 0.69777512550354, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bca6c4d4d804fef0bb6b8b5cb78cd2d965771a6493fd2d160f41370312166a59.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "bcc558e17f451a67e1dbe436e859c0ce9e3563392fa9d1c49cd662dba3819e90", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "759fc9e812ebb8decf2c7fd038a9b808df26de06c68a24a50931fac71b8e1f43", "generation_seconds_batch": 21.461706755915657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.759477972984314, "target_z": 0.8281249773019763, "ce": 5.804190158843994, "genuineness": 4.6203155517578125, "identity": 0.7166175842285156, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bcc558e17f451a67e1dbe436e859c0ce9e3563392fa9d1c49cd662dba3819e90.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "bd75b880b9252bdd317cec6113ecb810e6f463f37ec09962a80de7fadcd2f21f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "39e62988bf6b3609f965e4d413410f0066124ee436323df60c1a807e27a748de", "generation_seconds_batch": 14.931731909979135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1646086424589157, "target_z": -0.2871093830752584, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.34123656153678894, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bd75b880b9252bdd317cec6113ecb810e6f463f37ec09962a80de7fadcd2f21f.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "bd9b42d8559b62a868c313cd90e248050d8e0c1458975fc0f576853f3d6c2780", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "43446b9cffb9a024c71c8921d24d91b2b21afec45bad277cc25d504676175267", "generation_seconds_batch": 26.950108854100108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3182745575904846, "target_z": 0.000976587707901326, "ce": 6.006728172302246, "genuineness": 4.507347106933594, "identity": 0.6861653327941895, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bd9b42d8559b62a868c313cd90e248050d8e0c1458975fc0f576853f3d6c2780.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "bdafb2a350f90cd44d300e21d1fccb59ee50f339717b8e1a4d7c1d297ec85d7c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f5c484871fa1cfa9530c367415b7cd62f4379c72e10bd342aaf81e5f5f8edd83", "generation_seconds_batch": 20.721714751096442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4677732288837433, "target_z": 0.28124999650799637, "ce": 6.022613525390625, "genuineness": 4.250029563903809, "identity": 0.4853898882865906, "hypothesis": "I found the little box. It was beside the old window We Can Look Inside Together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bdafb2a350f90cd44d300e21d1fccb59ee50f339717b8e1a4d7c1d297ec85d7c.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "bdba89ba1122d42ba04b5eed74987b5804932bf4051d630121f32ebc64bf793d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "033d5ab38e135ed05eef9ee32d767406692b9dceaebbe34ffe86f8d33c7d3d12", "generation_seconds_batch": 21.117626238963567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6782174110412598, "target_z": 0.6757813176575707, "ce": 6.038498878479004, "genuineness": 4.7081804275512695, "identity": 0.49501848220825195, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bdba89ba1122d42ba04b5eed74987b5804932bf4051d630121f32ebc64bf793d.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "be88dea57fce6e4ec3bb0cbb1730d7a766f173a7a2fca49ed291878b00f3ff30", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 130, "seconds_generated": 10.4, "mp3_sha256": "c8ee6b99abefe63fa6da94932c9a50e38d358204d612cfb4cc16aa913b436bd9", "generation_seconds_batch": 21.911486500059254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4095629453659058, "target_z": 2.046875148410155, "ce": 5.307773113250732, "genuineness": 4.406930446624756, "identity": 0.7170249819755554, "hypothesis": "Die Tür ist noch offen. Ich habe dein Mantel auf den Stuhl gelegt. Sag mir.", "wer": 0.2631578947368421, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/be88dea57fce6e4ec3bb0cbb1730d7a766f173a7a2fca49ed291878b00f3ff30.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "bed36747ad8ee4e0882a445164a45579545e01e67365a7abb4f60899b7d55afb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4075d088d2faa6eb0068d135822282b1f3744e9e9219ae5cf1f367b6c6e5ef28", "generation_seconds_batch": 21.491338068968616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00312158465385437, "target_z": -0.6015624956349954, "ce": 6.078212261199951, "genuineness": 4.507347106933594, "identity": 0.5915117859840393, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bed36747ad8ee4e0882a445164a45579545e01e67365a7abb4f60899b7d55afb.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "bf58036d1de497ba17a66d1868f2abd89192a69bcb899c3fb99ea5617dc6e5dc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12683c731d5f274e130ef001309ff806d8e4098bc989b2247427ed854bf9b9db", "generation_seconds_batch": 21.468618754995987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11356031894683838, "target_z": -0.3828124921429918, "ce": 5.935244083404541, "genuineness": 4.049196243286133, "identity": 0.24643947184085846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bf58036d1de497ba17a66d1868f2abd89192a69bcb899c3fb99ea5617dc6e5dc.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "bf5a8847c4d3c8ebf1c356c8a039fe11263de2f679dc04cf6a64e842a1f8fac8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "50b320b44b4e675e35a256dc6264035e61d8421ff80b88d892a79b5cd65ba0a6", "generation_seconds_batch": 20.90115898195654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5334067940711975, "target_z": 0.40429686299623746, "ce": 5.863759994506836, "genuineness": 4.720732688903809, "identity": 0.5733502507209778, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/bf5a8847c4d3c8ebf1c356c8a039fe11263de2f679dc04cf6a64e842a1f8fac8.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "c0a790c7f93c82832006ee708f659617d6abda198d3d7182282d9b38dcbddd56", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "64edec1860ccb356f7944ac7e7baf2ef4371c9fa25491d486b970d706759b9c8", "generation_seconds_batch": 20.736993049038574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06355378031730652, "target_z": -0.47656250960301005, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.5738307237625122, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c0a790c7f93c82832006ee708f659617d6abda198d3d7182282d9b38dcbddd56.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "c0f08f4d8b922f67198598bb0a79ec5d7217ac8cce50b83a92bb532948492086", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2f157e3bc63f7dddddf3c908a34b1bab4d7785df77d80530e4a94498a79bb380", "generation_seconds_batch": 22.130511948955245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.138694167137146, "target_z": 1.5390625026190028, "ce": 5.708878040313721, "genuineness": 4.6203155517578125, "identity": 0.7041245698928833, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c0f08f4d8b922f67198598bb0a79ec5d7217ac8cce50b83a92bb532948492086.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "c11128aa96f34b92d0b9fc50211b613bcf223fbddc339e4c2e81d7e11e572d94", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fdc813820af58434d6d7f00d7a7911efeafb2186868b479bd497a04de510b825", "generation_seconds_batch": 22.04128434404265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08959884941577911, "target_z": -0.4277343813292566, "ce": 6.09409761428833, "genuineness": 4.833701133728027, "identity": 0.7607901096343994, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c11128aa96f34b92d0b9fc50211b613bcf223fbddc339e4c2e81d7e11e572d94.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "c13b94fe7c3ee087160955256ed29c8519b735767bd2972842239387e14902c0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b947a70200d33548ffc794dfa5e4871b81e86fde86ae4f71c32df8c106c7571b", "generation_seconds_batch": 24.3745497839991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393024444580078, "target_z": 3.890624690957677, "ce": 5.569881439208984, "genuineness": 3.9173994064331055, "identity": 0.0868222564458847, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c13b94fe7c3ee087160955256ed29c8519b735767bd2972842239387e14902c0.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "c1e35206e95e73d784e7081ae93a4cdf5fdbea775e77d6565412ac9e7a9da04b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "9c3bf6aa1c365943154d43710e76181393422c2b8c0830fa5c9f496708e36a62", "generation_seconds_batch": 21.86774013703689, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5762512683868408, "target_z": 2.3593749458739435, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.5563074350357056, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen het er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c1e35206e95e73d784e7081ae93a4cdf5fdbea775e77d6565412ac9e7a9da04b.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "c25af91bf82c6035cd2b3007508c9ab5d31f5d2185e3d9798c004a1170c49fb3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f35e20df7310e9574cc405c456b8702b114fae441a641eed952d73e0c27fa09a", "generation_seconds_batch": 21.079515608027577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1887006759643555, "target_z": 1.6328124642069626, "ce": 5.8875885009765625, "genuineness": 4.7332844734191895, "identity": 0.5929076671600342, "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/emo_Confusion/c25af91bf82c6035cd2b3007508c9ab5d31f5d2185e3d9798c004a1170c49fb3.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": "c30239487aea9fb69cb62bc36ef41cc20668775a7bf12f163ff4637cee023e97", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e54752e14e0395029718f9f67b9aea6c5cbafb0e9aa9f0618ba293d8d570ae66", "generation_seconds_batch": 22.031887800083496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0678515434265137, "target_z": 1.4062499266679234, "ce": 5.927301406860352, "genuineness": 3.9048471450805664, "identity": 0.39689379930496216, "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/emo_Confusion/c30239487aea9fb69cb62bc36ef41cc20668775a7bf12f163ff4637cee023e97.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": "c38a87b270daf8ec1ce24da2b5a26c122821b18515cd79991540194f9a339108", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "144c9d4147dd24f41d31f8f7db3ae093d066628a5c443e0f7f2fa28dca933887", "generation_seconds_batch": 21.85695292300079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.268008232116699, "target_z": 3.656249898731894, "ce": 5.895530700683594, "genuineness": 4.15588903427124, "identity": 0.43057864904403687, "hypothesis": "Niemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lurs terug.", "wer": 0.09523809523809523, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c38a87b270daf8ec1ce24da2b5a26c122821b18515cd79991540194f9a339108.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "c38b40ec0b85baf912775a695d22fee537acc60cd5306cd9c55a665e08853baf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1b588d1957758f64b7c88cba395a083ac225cc26c3bfed7f4e79a50d0e5815ab", "generation_seconds_batch": 21.106531191035174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33963149785995483, "target_z": 0.04101562216274704, "ce": 6.070269584655762, "genuineness": 4.331617832183838, "identity": 0.4381026029586792, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c38b40ec0b85baf912775a695d22fee537acc60cd5306cd9c55a665e08853baf.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "8a6c2f10c10bd8b39048cd21194f794b523eb79cbab3ace87d923bb85be580d5", "generation_seconds_batch": 15.310805763001554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.7065035104751587, "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/emo_Confusion/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "c3d75521bb7ecc9fa3e1d13441fa2aaa3b03362358044ea9cbc76c03058fa6f2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d99ab62f7554b88acca1050295e94813a244df0df8fe22f670db50818150841c", "generation_seconds_batch": 21.92049885704182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0958496630191803, "target_z": -0.41601563613076165, "ce": 5.990842819213867, "genuineness": 4.7332844734191895, "identity": 0.7510961294174194, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c3d75521bb7ecc9fa3e1d13441fa2aaa3b03362358044ea9cbc76c03058fa6f2.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "30846d4c056bbd588841be3bcbf785649b4d45e1a1adae4e190065d1ec849537", "generation_seconds_batch": 15.53886979399249, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11981113255023956, "target_z": -0.3710937469444968, "ce": 6.268836498260498, "genuineness": 4.287685871124268, "identity": 0.7774715423583984, "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/emo_Confusion/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "c4974fc704422ba7fd2b8a7562cb6edffbf0671fb03c235580b58f39290fb09e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6955cec0048061c2a77e997f0f44864c183acf2ed5f4a41f46c4f14c14199045", "generation_seconds_batch": 19.78072325210087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1302291601896286, "target_z": -0.3515624956349954, "ce": 6.038498878479004, "genuineness": 4.557555198669434, "identity": 0.7888255715370178, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c4974fc704422ba7fd2b8a7562cb6edffbf0671fb03c235580b58f39290fb09e.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "c4baa02c8d1c29cf6db64e28257adf7234a8599da78276c35d5131bbcc923549", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "122b410292eb015c780a1087635005fbc71420fc4bfd06395953481ac4aa94b5", "generation_seconds_batch": 19.314188722986728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15314881503582, "target_z": -0.3085937539285041, "ce": 6.022613525390625, "genuineness": 4.6203155517578125, "identity": 0.6386777758598328, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c4baa02c8d1c29cf6db64e28257adf7234a8599da78276c35d5131bbcc923549.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "c510ce600066b4c1e65e20afed430a06eb1f74f3f7d8b0950f27b8934d37de5f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1a0a5e5c83b7aa1ed1a2337009b15853c3d31bb0ee3f9b95d02cdeb98d083b36", "generation_seconds_batch": 20.664601166034117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 5.95112943649292, "genuineness": 4.607763767242432, "identity": 0.45491644740104675, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c510ce600066b4c1e65e20afed430a06eb1f74f3f7d8b0950f27b8934d37de5f.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "c6585b5fb62fb0bbb4d7cd262a7b97ac0eeaab4c7ced02b81511c932a39dbf66", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "843f0cd1849e09984f3a5d5e4aa8efe8c1538fc2d8f40edad1c1c21a7b08ab6b", "generation_seconds_batch": 19.809968107030727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03646692633628845, "target_z": -0.5273437294844786, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.5969491600990295, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c6585b5fb62fb0bbb4d7cd262a7b97ac0eeaab4c7ced02b81511c932a39dbf66.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": "c6623cbf2f1ec0ba7c18dcfffcd1398b527ba3eeafa94f8c34034c9c95837ee7", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4aad6ebb35d8501749b83d85a7f3e53018591e0ee0f23a9f831ba2d52fc53595", "generation_seconds_batch": 29.14194281003438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.984637975692749, "target_z": 3.125000041904044, "ce": 5.649308204650879, "genuineness": 4.319066047668457, "identity": 0.6385716199874878, "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/emo_Confusion/c6623cbf2f1ec0ba7c18dcfffcd1398b527ba3eeafa94f8c34034c9c95837ee7.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "c690acbb9e14efa4fde753c39582140d88bb2e2f26903e3fd3ee6d301df0e5a0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a42bebcf3d24e7dd4c8f6334b9f38b212e957685cf35e91ceeca9cbd48636518", "generation_seconds_batch": 22.34608659008518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.5800938606262207, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c690acbb9e14efa4fde753c39582140d88bb2e2f26903e3fd3ee6d301df0e5a0.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": "c69b4e83472d3ac43fe9d698abba6f8d10eeb73f4fe2295824c20319046b2736", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f882111e02bab8cd8462041a9d60a62eba3ef6a172b6faf708480f9c4546ddb4", "generation_seconds_batch": 16.361110532074235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8512872457504272, "target_z": 2.875000069840073, "ce": 5.839932441711426, "genuineness": 4.306513786315918, "identity": 0.44576582312583923, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed. En controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c69b4e83472d3ac43fe9d698abba6f8d10eeb73f4fe2295824c20319046b2736.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "c6c2781553fe3e1d5f9b44a738b9df70267cdf390afe3c1f188d8af56edda95f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9cf1d245afc62f08cddbef415bf5d2a8630e687d8368d3e631a0485be1b84c52", "generation_seconds_batch": 20.238970909966156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9261664152145386, "target_z": 1.1406249982539982, "ce": 6.006728172302246, "genuineness": 4.457139015197754, "identity": 0.33865803480148315, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c6c2781553fe3e1d5f9b44a738b9df70267cdf390afe3c1f188d8af56edda95f.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "c79fa059e03d2a784cf2b8c58ad591c843c4a84878f994939f6e8ffebfa9ef0e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "5fe69b2576e28090b2a031f33bc22f7674a27bb199eac6666548508cfeb47a5e", "generation_seconds_batch": 20.342482059961185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.284677028656006, "target_z": 3.687499811431803, "ce": 5.95112943649292, "genuineness": 4.344170093536377, "identity": 0.5418469309806824, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c79fa059e03d2a784cf2b8c58ad591c843c4a84878f994939f6e8ffebfa9ef0e.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "c41ea975d8a8736b7d367a2abdbc78c2b571219a2763ff3d9b4f17f928d177d0", "generation_seconds_batch": 15.878778078011237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6865518093109131, "target_z": 0.6914062740075251, "ce": 6.300607204437256, "genuineness": 4.293961524963379, "identity": 0.7681465148925781, "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/emo_Confusion/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "c882559bdeb0d47fad3c3fddcd5a2cb65006729f7d9e88e6aa5b2329d8bb4da1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "cc9a98b0e071430beed6b6064a5e5f46040efbef846406638ccd6889e812ae92", "generation_seconds_batch": 21.294460720033385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03646692633628845, "target_z": -0.5273437294844786, "ce": 6.054384231567383, "genuineness": 4.607763767242432, "identity": 0.6085445880889893, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c882559bdeb0d47fad3c3fddcd5a2cb65006729f7d9e88e6aa5b2329d8bb4da1.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "b0b9a5713b8a3eb797d55e9fa648eb61d40424383f6e0c0a050f24920f60a620", "generation_seconds_batch": 15.965634776977822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0698046088218689, "target_z": -0.46484373646848587, "ce": 6.443574905395508, "genuineness": 4.331617832183838, "identity": 0.24774269759655, "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/emo_Confusion/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c8ab5088246d98b06bdd58d179f882f1330fb491fa8ebb03d2bde23029338526", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b8ee28cd65557166d4cab84dcf0335055af237b35304e5381130677b2748e55a", "generation_seconds_batch": 21.85695292300079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0513134002685547, "target_z": 3.2500001396801457, "ce": 5.982900142669678, "genuineness": 4.118232727050781, "identity": 0.41117459535598755, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c8ab5088246d98b06bdd58d179f882f1330fb491fa8ebb03d2bde23029338526.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": "c8cd4ac63e0e0173500e114a38c91e8fd344b2fbb283ace2139d3375a384e865", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7f5e6bf762c315ae522a633c0c4983db60d296b2d015aae09a0cdf7cbbfd0e5", "generation_seconds_batch": 20.89111759501975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293011426925659, "target_z": 3.7031247677817576, "ce": 5.748591423034668, "genuineness": 4.557555198669434, "identity": 0.7745805978775024, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich lass sie zweimal, dann leg dich das Handy weg. Sag mir, wann wir reden können.", "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/emo_Confusion/c8cd4ac63e0e0173500e114a38c91e8fd344b2fbb283ace2139d3375a384e865.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "c8e29ead5815793d4dbcfdf2dc5274f316c90b64fb8af87940f247dd81141ac5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d2f814075040552477f0da9a82c4b101a7eaf0caae4b697041397f7e7894549f", "generation_seconds_batch": 19.314188722986728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2281586080789566, "target_z": -0.16796875567450592, "ce": 6.0146708488464355, "genuineness": 4.7081804275512695, "identity": 0.44616395235061646, "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/emo_Confusion/c8e29ead5815793d4dbcfdf2dc5274f316c90b64fb8af87940f247dd81141ac5.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "c92175905ab70c9a166712f8dccce0b8e98f026f85f09fa1fc7e45497bd409bd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "941397200966c9723c674cc4dab925e44ec1401ec1d71926aec1fa3d7bbf7a3e", "generation_seconds_batch": 15.966613859985955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29196903109550476, "target_z": -0.04833983441980275, "ce": 5.804190158843994, "genuineness": 4.833701133728027, "identity": 0.7748534679412842, "hypothesis": "He encontrado la cajita. Estaba junto a 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/emo_Confusion/c92175905ab70c9a166712f8dccce0b8e98f026f85f09fa1fc7e45497bd409bd.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": "c9b50ccf9b973ed43f976fd05527b5087497367f1eee1f60911db823681e8c0c", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f9e0eb8ea05c3a970ccf7895fd1a7e0b1a423db8b94b9c3851baa3a72fbd69b7", "generation_seconds_batch": 18.491003404953517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9513002634048462, "target_z": 3.062499993015993, "ce": 5.7406487464904785, "genuineness": 4.293961524963379, "identity": 0.6552382707595825, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c9b50ccf9b973ed43f976fd05527b5087497367f1eee1f60911db823681e8c0c.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "c9c88581f66e97fbffadf36292b7688870a3a0addbc5cfb68139f970a29e0c35", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "fb1b1fd8c49854e620a467d02d260fc895753370ac363e4768011ef54d099472", "generation_seconds_batch": 15.810208895942196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24899466335773468, "target_z": -0.12890625305550318, "ce": 5.772419452667236, "genuineness": 4.281409740447998, "identity": 0.22325125336647034, "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/emo_Confusion/c9c88581f66e97fbffadf36292b7688870a3a0addbc5cfb68139f970a29e0c35.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "c9cb1e158c187955adc606328ff73a68115025a09b19a8ad45648d41b6a423a0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "3b955c6015042556867bb82b43a64cb26f95ef13f02b161fe7ea2fa42ed759a3", "generation_seconds_batch": 21.53755801601801, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7032206058502197, "target_z": 0.7226561867074339, "ce": 5.6254801750183105, "genuineness": 4.607763767242432, "identity": 0.7280131578445435, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/c9cb1e158c187955adc606328ff73a68115025a09b19a8ad45648d41b6a423a0.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "ca6358dc16a9928088316ebfc7de055dc59318855aa61ebc259f5cd1deeb782b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d44fbb2bea715e3bb16374cdcfaa9d271c0d5b7058b6cccffc869fb8f1625fcd", "generation_seconds_batch": 15.781815703958273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9928418397903442, "target_z": 1.2656250960301003, "ce": 6.173524379730225, "genuineness": 4.808597087860107, "identity": 0.6732059121131897, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ca6358dc16a9928088316ebfc7de055dc59318855aa61ebc259f5cd1deeb782b.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": "ca9cadd434e95178e6cf14b2a95033124e46f1bb49d8e7849cce988d59f04d03", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Someone moved the little blue chair.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4beb784faa1f219a8d80556b22685664860919ff7451722396ea9af12db969db", "generation_seconds_batch": 15.9259085690137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24899466335773468, "target_z": -0.12890625305550318, "ce": 5.95112943649292, "genuineness": 4.143336772918701, "identity": 0.22464734315872192, "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/emo_Confusion/ca9cadd434e95178e6cf14b2a95033124e46f1bb49d8e7849cce988d59f04d03.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "cab930c4bf33f56394cb31358996dbaf3c2b452eccf83237b80fe5c4f4bbae04", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "2ebbf24a71d25af5135da8684c7a89fb64cf294ab9e71a5b7355775ce6d75a21", "generation_seconds_batch": 21.352340943994932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6969698667526245, "target_z": 0.7109375811890848, "ce": 5.6413655281066895, "genuineness": 4.168440818786621, "identity": 0.43109142780303955, "hypothesis": "The door is still open. I I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cab930c4bf33f56394cb31358996dbaf3c2b452eccf83237b80fe5c4f4bbae04.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "cabedaef8824a7eafec99930910c01364a6b1baa0c0a55033a881deb3a1466df", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "126cff3bfda4a8e786591c669ac351c923ad07285630e31ca6147548322f4b56", "generation_seconds_batch": 21.773904109024443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4845726490020752, "target_z": 2.187499979047978, "ce": 5.689021587371826, "genuineness": 4.356721878051758, "identity": 0.7471166849136353, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cabedaef8824a7eafec99930910c01364a6b1baa0c0a55033a881deb3a1466df.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "cb54c82105ac2dc3c03c93f05c31a521116239498eca8fd3e7165c9db305b919", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "094c158fa8dbb83f18117aac31e8678157dd9dde554ddb16dea72afaa46eb116", "generation_seconds_batch": 15.826243589050137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.577162504196167, "target_z": 0.48632810732173154, "ce": 5.788304805755615, "genuineness": 4.570107460021973, "identity": 0.7300105094909668, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cb54c82105ac2dc3c03c93f05c31a521116239498eca8fd3e7165c9db305b919.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "cbcc30aa8d95d3e51734498a5bffa2f0d651a2a50cb3760640efd2e8a11fcc6d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "215c7b0136b8601aa5a2e571abfbd6c68f00ead41f6299fa1d5149cb896daf9a", "generation_seconds_batch": 18.816177662927657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6365453004837036, "target_z": 0.5976563124195652, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.5380151271820068, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cbcc30aa8d95d3e51734498a5bffa2f0d651a2a50cb3760640efd2e8a11fcc6d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "cbd1a4056845ca3e2ec3592b9d1d366380fab86b2a982301101e0c2fa35d2a17", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8a373110ff2dec9d73556f68b3e7042ed2e90d62cc858f50cbd6e07cc70d7504", "generation_seconds_batch": 20.606669208966196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06459558010101318, "target_z": -0.4746093900592657, "ce": 5.863759994506836, "genuineness": 4.68307638168335, "identity": 0.6323037147521973, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cbd1a4056845ca3e2ec3592b9d1d366380fab86b2a982301101e0c2fa35d2a17.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "cc81b62f19fedc00fb27f346dcd3a471e9ce775e814aab2c821162d9f7646975", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a5ba9c8c1912311aff5567234629aef821060477ea8bdaf79cfc48b7aa26085f", "generation_seconds_batch": 22.874549507978372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0178450345993042, "target_z": 1.3124999650799636, "ce": 5.272031307220459, "genuineness": 4.482243061065674, "identity": 0.49008581042289734, "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/emo_Confusion/cc81b62f19fedc00fb27f346dcd3a471e9ce775e814aab2c821162d9f7646975.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "ccb55aae13b752d4a7a7f1a22670a88d25fb906fd0966ab5ee830c41566d2cf2", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "570b1d3959ef26b4bc31d4451a2cd4b6b5f5b7cfa53a4c3d325e71dcf855523c", "generation_seconds_batch": 15.829274791991338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22711680829524994, "target_z": -0.16992187521825022, "ce": 6.229123115539551, "genuineness": 4.545003414154053, "identity": 0.4696584939956665, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ccb55aae13b752d4a7a7f1a22670a88d25fb906fd0966ab5ee830c41566d2cf2.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "ccd9b5a25017c6165074c6cb2d3874d7cbe127789b64ccd8f9fbb1691d74c47a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "44fadef4d8898c9b8d531188d77344453772e6d24d782d01345a0b8a589a90c7", "generation_seconds_batch": 20.465676228981465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09793327748775482, "target_z": -0.41210936910724383, "ce": 5.847874641418457, "genuineness": 4.293961524963379, "identity": 0.6117942333221436, "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/emo_Confusion/ccd9b5a25017c6165074c6cb2d3874d7cbe127789b64ccd8f9fbb1691d74c47a.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "ccf638b3e3fc700b008b9a15b7ce6d210b2bcd1bf55614603766d7e5518acbda", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a34da8c3c0cf232031e01e688d760fe7fb40a647af026ce88ead7def0385e13a", "generation_seconds_batch": 16.31303141801618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7262709140777588, "target_z": 2.6406250541260565, "ce": 5.442798614501953, "genuineness": 4.344170093536377, "identity": 0.6340360045433044, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ccf638b3e3fc700b008b9a15b7ce6d210b2bcd1bf55614603766d7e5518acbda.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "cd7611f5b42515b9d8402bed3c235ed7d14f85777d222c8b48b09d6802438f6c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "a4241b305856fb4fab44914c4286b0248a6f5b5f65f8d5b2b6d58e36e10d887a", "generation_seconds_batch": 21.778996143024415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41151589155197144, "target_z": 0.17578126178551232, "ce": 5.784333229064941, "genuineness": 4.7332844734191895, "identity": 0.5798467397689819, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cd7611f5b42515b9d8402bed3c235ed7d14f85777d222c8b48b09d6802438f6c.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "cd8f223fc88606eba73a10e7972dd79c128d13e6e1bb1e396289da240a355db3", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "3a057ee4895b2eb108e2a40fb0768aeb182cc4fcbcbc35e3edff0ebab1cedbe6", "generation_seconds_batch": 21.218752069980837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0843898355960846, "target_z": -0.4375000069840073, "ce": 6.086154937744141, "genuineness": 4.15588903427124, "identity": 0.3548924922943115, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cd8f223fc88606eba73a10e7972dd79c128d13e6e1bb1e396289da240a355db3.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "cda195c116be18e61f860512a4c3fc9a2d191015284b3ac6c20e66d9f5145436", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "03d7dff4bbef8442863d2bbdb8c517973c73fef320fb13a29d13dd914b86e1de", "generation_seconds_batch": 20.015073664020747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05105215311050415, "target_z": -0.5, "ce": 6.09409761428833, "genuineness": 4.406930446624756, "identity": 0.398773729801178, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cda195c116be18e61f860512a4c3fc9a2d191015284b3ac6c20e66d9f5145436.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "ce6c277bd2d356896d83010dac56aba7ff2d1a429310e3ed7236eb8d447254aa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bb3ad61e1841ffad299fb72a0df9026c4e8aff037d2fe6089b9f988a37c12253", "generation_seconds_batch": 14.834789806045592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42297571897506714, "target_z": 0.197265632638758, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.4149012863636017, "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/emo_Confusion/ce6c277bd2d356896d83010dac56aba7ff2d1a429310e3ed7236eb8d447254aa.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": "ce8f7f81739e0944d1b57dbf596c9c250a10ed046389704f95b837d0b9186230", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0720cdeb6fa8da8698e258d90b6dc913d115258433593e5748bede578b9124c8", "generation_seconds_batch": 16.27003201493062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2763426303863525, "target_z": 3.6718748550818487, "ce": 5.677107334136963, "genuineness": 4.13078498840332, "identity": 0.6940601468086243, "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/emo_Confusion/ce8f7f81739e0944d1b57dbf596c9c250a10ed046389704f95b837d0b9186230.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "cf2b737414a6447931826a0405bccd8990a390852485504642cc1fd69a85797a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bc7e666546623cc8d24cde8b2f5832da8f7668577d2559cc06cc7fdf95b7ad6f", "generation_seconds_batch": 21.690407857997343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41985028982162476, "target_z": 0.1914062181354667, "ce": 5.77639102935791, "genuineness": 4.645420074462891, "identity": 0.5129204392433167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cf2b737414a6447931826a0405bccd8990a390852485504642cc1fd69a85797a.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "17f29d1b109c83f407aa858bc1214ab2b5c1943fa5fb9ce7d34cc24cdd0d3096", "generation_seconds_batch": 14.662117762025446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.056261152029037476, "target_z": -0.4902344022812785, "ce": 6.41180419921875, "genuineness": 4.582659721374512, "identity": 0.5851061344146729, "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/emo_Confusion/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "cf9f7fd2db5cdaae68e4a5fc0ad5ba0831d11a4b7211acad1e11fcbbf79fe580", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f70447e874271f5fad5cd5c911d1cb1381c4d2a717953085ed91002e89042722", "generation_seconds_batch": 14.121279977960512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5156961679458618, "target_z": 0.371093774880526, "ce": 5.863759994506836, "genuineness": 4.344170093536377, "identity": 0.37872153520584106, "hypothesis": "The door is till open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cf9f7fd2db5cdaae68e4a5fc0ad5ba0831d11a4b7211acad1e11fcbbf79fe580.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "cfab5c93a140465181adbef2c43e84b53c16e23316d0a9a27f8069bfe9f6f274", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2fe15e22b89747fe8e4b346bfbf75d14e5765a8356611283aba0745370ee013d", "generation_seconds_batch": 20.450848050997593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2012022733688354, "target_z": 1.6562498987318943, "ce": 5.550024509429932, "genuineness": 3.973883628845215, "identity": 0.27604442834854126, "hypothesis": "Ik heb het een klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in komen.", "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/emo_Confusion/cfab5c93a140465181adbef2c43e84b53c16e23316d0a9a27f8069bfe9f6f274.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": "cfded47db350fa46137d4c622ca5d73b16d5912435887276c32c5966e7b712bb", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2dbabb52a24c7cfb9542ce45ac7fcf32c9c2519e92f656899a0682dd11280af9", "generation_seconds_batch": 21.344913888024166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718066930770874, "target_z": 4.499999776511767, "ce": 5.700935363769531, "genuineness": 4.20609712600708, "identity": 0.4482015073299408, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de l en daarna naast het raam neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cfded47db350fa46137d4c622ca5d73b16d5912435887276c32c5966e7b712bb.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "cfeb92061bc7ab488670c6d68906152a172a6514acd1a77a082526e5c1d48dd4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6ca419b338edd9ad7939efebc957e24fdf01ceb8a7f319fb91437969f3bbb259", "generation_seconds_batch": 22.26091180893127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04480132460594177, "target_z": -0.5117187731345242, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.6442362666130066, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/cfeb92061bc7ab488670c6d68906152a172a6514acd1a77a082526e5c1d48dd4.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "d023f4746078477febb1b6f39466e6dc39c09adb5416506884aec7a8c874738e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e9e70974a65b1d0137b8d38f5ccbb1ba2b66ac6c9b6b429482771a82eeee7865", "generation_seconds_batch": 14.85682431096211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7990664839744568, "target_z": 0.9023437434524931, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.564105749130249, "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/emo_Confusion/d023f4746078477febb1b6f39466e6dc39c09adb5416506884aec7a8c874738e.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "d03294ad01754defd37efbbfffbc2ac280b62322686cde0a1d9fa561b4377d61", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "cf080509ad16ea176f46590c20ceb8e49b99a33260e41910fc077113b6edf22b", "generation_seconds_batch": 24.3745497839991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293011426925659, "target_z": 3.7031247677817576, "ce": 5.692992687225342, "genuineness": 4.105680465698242, "identity": 0.30699557065963745, "hypothesis": "De deur staat nog open. Ik ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d03294ad01754defd37efbbfffbc2ac280b62322686cde0a1d9fa561b4377d61.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "d04034aa93c56d5ce0947128c792eab7f8b0f3b3a7484e43f97381bc3be7babd", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e6aaeaa673d5266864769be2d0391033069c35dc37cd62351610043bb6b4f875", "generation_seconds_batch": 15.530873834970407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.368021249771118, "target_z": 3.843749821907814, "ce": 6.070269584655762, "genuineness": 4.381826400756836, "identity": 0.7312185764312744, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d04034aa93c56d5ce0947128c792eab7f8b0f3b3a7484e43f97381bc3be7babd.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "d0b4552dd77f40cf79a72f08791c47baa05b935cff462debb789128aa8c597d1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4a7f103193e41fc7d05ff7e9ee4fb6a04e1ba274ac3388435cb279603ef0f90b", "generation_seconds_batch": 15.126584465033375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31046101450920105, "target_z": -0.013671892678268462, "ce": 6.0305562019348145, "genuineness": 4.532451152801514, "identity": 0.6905227899551392, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d0b4552dd77f40cf79a72f08791c47baa05b935cff462debb789128aa8c597d1.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "b512d1f7730607fd72a8752b934f8c734e72f39cecd0c29714fbbbb910197d57", "generation_seconds_batch": 16.2280199440429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.3641486167907715, "genuineness": 3.860914945602417, "identity": 0.18058045208454132, "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/emo_Confusion/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "d1b35e1ac40999aceaf5de9dc8ec66b090c71858a760bb091dc4d111a1f98bb8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1f31f3c95bc80d96fb275251e54af0400340af3abfac77714a85589147546dd8", "generation_seconds_batch": 19.042343620909378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2470417022705078, "target_z": 1.7421876056331103, "ce": 5.252174377441406, "genuineness": 4.632867813110352, "identity": 0.5024720430374146, "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/emo_Confusion/d1b35e1ac40999aceaf5de9dc8ec66b090c71858a760bb091dc4d111a1f98bb8.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "d1d8e2f300274731d791f9dc46736ba1113b66f625a5c66fe85d48e4ded3c36f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "5304459d0e2caf0390ab6ae7dd0e56fc8254f8b24240303d463459a6fafb5e04", "generation_seconds_batch": 22.397199576022103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1073095053434372, "target_z": -0.39453123734148676, "ce": 6.109982967376709, "genuineness": 4.250029563903809, "identity": 0.48043376207351685, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d1d8e2f300274731d791f9dc46736ba1113b66f625a5c66fe85d48e4ded3c36f.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "d1e514196b6a8516ebf38d37c51a39884b50a906d5ad35f7d939bf565f0fb407", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ef913ad0dc0aad4927a4d0664c7e9500cad23b574b167fcfd9eeea20363fe3a3", "generation_seconds_batch": 20.643940870068036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.1894097328186035, "genuineness": 4.7081804275512695, "identity": 0.5685620307922363, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d1e514196b6a8516ebf38d37c51a39884b50a906d5ad35f7d939bf565f0fb407.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "d81af0220877bcda0c4116f1aa4f486d833248426d54daac1c940aa28e71c142", "generation_seconds_batch": 15.820263803005219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05834478139877319, "target_z": -0.48632810732173154, "ce": 6.109982967376709, "genuineness": 4.425758361816406, "identity": 0.8047922849655151, "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/emo_Confusion/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "d238685d1cf6e5ac2ffc80b7c919e01fabfbff7d55d56b63f8b7a555f1dcff2c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0439db26e4cb5ae1186404f95bace64ee21df1bbaf4107eb2effee7bd3effef3", "generation_seconds_batch": 21.017572115990333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9553369283676147, "target_z": 1.195312568967072, "ce": 6.165581703186035, "genuineness": 4.419482707977295, "identity": 0.5100909471511841, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d238685d1cf6e5ac2ffc80b7c919e01fabfbff7d55d56b63f8b7a555f1dcff2c.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d28d8e90f9dc03033895bc00ca0326f49017d10b497d653b80d8358bd6f7dab1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3a281dbae8dffab66229bdb0087382ef2c8b330c8d3673f1c82d4f2ae08defcf", "generation_seconds_batch": 19.655113793094642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2893645167350769, "target_z": -0.05322266121519269, "ce": 5.8875885009765625, "genuineness": 4.281409740447998, "identity": 0.7376810312271118, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d28d8e90f9dc03033895bc00ca0326f49017d10b497d653b80d8358bd6f7dab1.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "d3332ab94ff8c8de58f621bb0ef91f085ecdceb9454e4b6544834a56cb30ab15", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "14a97e3c89ed6082f4584a563530119cbda58cb039d2af2d0b153b68bc629b98", "generation_seconds_batch": 23.630956389941275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0513134002685547, "target_z": 3.2500001396801457, "ce": 6.0305562019348145, "genuineness": 4.356721878051758, "identity": 0.4499703645706177, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d3332ab94ff8c8de58f621bb0ef91f085ecdceb9454e4b6544834a56cb30ab15.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "d354840d8d6bf4cc981ef0b57d7eca76a79b481473048f7fa4496f6de3e77080", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4a2d5cf73038df17295eeb74e191b3b7932f78425864498a2fd22d9aa6874365", "generation_seconds_batch": 16.004264255054295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021881669759750366, "target_z": -0.5546875148410155, "ce": 5.95112943649292, "genuineness": 4.6956281661987305, "identity": 0.710442841053009, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d354840d8d6bf4cc981ef0b57d7eca76a79b481473048f7fa4496f6de3e77080.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "d38c70c630d5d0f03cfce938e140fa65a9b2c74d79dc8c02b82f74c2889abe50", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dba7a94ba5bfa1f096cd8acb407f9201ebbbae17d8feb29a1902bd274a8ce864", "generation_seconds_batch": 20.611866575898603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5679168701171875, "target_z": 2.3437499895239893, "ce": 5.4666266441345215, "genuineness": 4.18099308013916, "identity": 0.6890243291854858, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d38c70c630d5d0f03cfce938e140fa65a9b2c74d79dc8c02b82f74c2889abe50.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "d3f8cf58a0bc9dd0495696a4e5e34fe367bf30162f1ac23af481668080da178c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1908df5f5c91666386f4b85cda939bbf9accee435a145dc98fc2d6be8b1905fe", "generation_seconds_batch": 22.573375951964408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.138694167137146, "target_z": 1.5390625026190028, "ce": 5.824047088623047, "genuineness": 4.494794845581055, "identity": 0.5694658160209656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d3f8cf58a0bc9dd0495696a4e5e34fe367bf30162f1ac23af481668080da178c.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "3db192e452a2b6693f0c3201d01810f78ea21f6ec6f8e590fb64f581f80ae042", "generation_seconds_batch": 15.678969554952346, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.117925643920898, "genuineness": 4.607763767242432, "identity": 0.769536554813385, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis le bien et et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "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/emo_Confusion/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "d448ab987576fc43092f88498080592edb2838795971c82b6382031560f07383", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3efcfb138983b95ce6699ffc59af41b35d1cbf577f1723b9613b70d627094dd4", "generation_seconds_batch": 20.714682637015358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4552716016769409, "target_z": 0.25781250611100637, "ce": 6.054384231567383, "genuineness": 4.281409740447998, "identity": 0.48852214217185974, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d448ab987576fc43092f88498080592edb2838795971c82b6382031560f07383.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d49288fa81d8e3af9fc6a2dc9a4c47abea46ba78c518f157fad4bc6bfbc17007", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4caa4b3a97b8fdc99248c64a9bce20966042abb65cea6eab974cca0caf113dc6", "generation_seconds_batch": 20.391116392915137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16148324310779572, "target_z": -0.29296874170649134, "ce": 5.927301406860352, "genuineness": 4.545003414154053, "identity": 0.6109594106674194, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d49288fa81d8e3af9fc6a2dc9a4c47abea46ba78c518f157fad4bc6bfbc17007.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "d516539309d8e2cb63dd5e07e38551c737c95f4d6dda4dabdb59f36a35ed0080", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "afb570dfbf47e44ec9a2eba34623c74f6f372055a1faaa1abd4c9a64720e98e8", "generation_seconds_batch": 21.46161303005647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.88462495803833, "target_z": 2.9375001187281238, "ce": 6.0305562019348145, "genuineness": 4.469690799713135, "identity": 0.6754813194274902, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d516539309d8e2cb63dd5e07e38551c737c95f4d6dda4dabdb59f36a35ed0080.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d56722df232d5c568a81717eec935d460bf7766b9b6f0c92c37b2a150871f123", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "350c9790dbf9c1d9157b7221b3db7fc1ef6afba39ec7c0eec01622c51aaf6541", "generation_seconds_batch": 20.108450368978083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3595563173294067, "target_z": 1.9531249633339618, "ce": 5.613565921783447, "genuineness": 4.796044826507568, "identity": 0.6069495677947998, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d56722df232d5c568a81717eec935d460bf7766b9b6f0c92c37b2a150871f123.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "d5f9190afb0475be101ba3783cc68c5b005adec3eaa9cc83f61d71122b0e67b9", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d031778eebe61fa878f076968719e0a1288a44f7d6358205fd00223e0a7df3b5", "generation_seconds_batch": 14.70974964601919, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0761860609054565, "target_z": 1.4218751065061113, "ce": 5.927301406860352, "genuineness": 4.262581825256348, "identity": 0.7483031749725342, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d5f9190afb0475be101ba3783cc68c5b005adec3eaa9cc83f61d71122b0e67b9.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "d60391c33200330e42902d39aa90dca6185aa0996b7eb8f10c1468b9273a0937", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3c71016654d1ccf7d7e1fc5fc90d5eea844b1e435d20be65b446ef391a684249", "generation_seconds_batch": 16.503352447994985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38234540820121765, "target_z": 0.1210937469444968, "ce": 5.593709468841553, "genuineness": 4.595211505889893, "identity": 0.7484124302864075, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d60391c33200330e42902d39aa90dca6185aa0996b7eb8f10c1468b9273a0937.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d608445388fef81fc6247b08dd25594368a4a3ec14253cdd0e4cb8bffefce5af", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "85feb1eae284195edbfc8ead7a44565409cc1844440122c50501d4c328fdbb8e", "generation_seconds_batch": 21.789721698034555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6429266929626465, "target_z": 2.484375043650046, "ce": 5.657250881195068, "genuineness": 4.337893962860107, "identity": 0.6606663465499878, "hypothesis": "La porte La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d608445388fef81fc6247b08dd25594368a4a3ec14253cdd0e4cb8bffefce5af.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d63f73edb4a8ec3dbd6c4127263c863534601c6ae830fb065825b017849b0191", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e02b0ddf9ef48d67fd05f4762b5ed672579db74a8b828273f445654176bed2c0", "generation_seconds_batch": 15.814881471917033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27451884746551514, "target_z": -0.08105465661759274, "ce": 5.772419452667236, "genuineness": 4.68307638168335, "identity": 0.7678745985031128, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d63f73edb4a8ec3dbd6c4127263c863534601c6ae830fb065825b017849b0191.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "d6848b5de83b25f0dd00b4d42230f176f8f25966d00c8d70a10e76e4b2519f2f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e215fbed1fa5b460ebf10cb724ca0f28f67ec7bb876e74a9b180df947ed3dcc0", "generation_seconds_batch": 21.195637193974108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7429397106170654, "target_z": 2.6718749668259654, "ce": 6.054384231567383, "genuineness": 4.293961524963379, "identity": 0.4187270700931549, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d6848b5de83b25f0dd00b4d42230f176f8f25966d00c8d70a10e76e4b2519f2f.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "d6ae4a8ec63331bfedfbd1683f4e32b5e49125b3ab2dc76dfbc6567c57d9a61b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c2b66205454f58dafea2f3dbaffe6c89d5e678c84cedbba239b77a3945c7ddde", "generation_seconds_batch": 20.010698296013288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6427960395812988, "target_z": 0.6093749179379143, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.6591488122940063, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d6ae4a8ec63331bfedfbd1683f4e32b5e49125b3ab2dc76dfbc6567c57d9a61b.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "d884ce3f1864fad7890b4597cf526eff5c4f6dfc97b57fe312c266c73c8df107", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1d810e0ec0c86dca29b2c8ff81c8224be7226973f721a3c08f88118188c4fa06", "generation_seconds_batch": 22.060853286995552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8594909906387329, "target_z": 1.015624900477896, "ce": 6.1020402908325195, "genuineness": 4.400654315948486, "identity": 0.5562374591827393, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d884ce3f1864fad7890b4597cf526eff5c4f6dfc97b57fe312c266c73c8df107.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "d8a3bbb29b439b2764ec97d7ecfc96097ed40e889f479bf7ad706dab564834ab", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8be42f6a352ecc1774431a2f51723201dcccd7126942fcc3f8a2fea58f7ed9b4", "generation_seconds_batch": 21.778996143024415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3422360122203827, "target_z": 0.04589844895813697, "ce": 6.0146708488464355, "genuineness": 4.6956281661987305, "identity": 0.7298353910446167, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d8a3bbb29b439b2764ec97d7ecfc96097ed40e889f479bf7ad706dab564834ab.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "d8e01afd015bbaf7dc05007532766e38a7d42e6d8c3647fa4e8bff4898309c6a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bdce699d9845cd27db19c90c6900435d5eaca7c74da6f7e20058208dfdb30b4", "generation_seconds_batch": 20.3413369449554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06667917966842651, "target_z": -0.4707031509717771, "ce": 6.0305562019348145, "genuineness": 4.13078498840332, "identity": 0.42479950189590454, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d8e01afd015bbaf7dc05007532766e38a7d42e6d8c3647fa4e8bff4898309c6a.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "d8e07b4ac2153e3075119ade154e58b2479b380683a01020b77dc90c5705cd28", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "354d4f0e04babfa1a51ab2971b3a6bb6e32928f8a182de690eee2a0f3c5a57c3", "generation_seconds_batch": 20.288248706026934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03855052590370178, "target_z": -0.52343749039699, "ce": 6.109982967376709, "genuineness": 4.570107460021973, "identity": 0.7358217239379883, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d8e07b4ac2153e3075119ade154e58b2479b380683a01020b77dc90c5705cd28.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "d8f1644681b620e9b22ecad74743d2a06cfd4ae17fbe27f9762d165719fdcfc1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "fe13366acd6a13581fa82a966bb434c06d37e9b68a31471b9634a574829b69e5", "generation_seconds_batch": 20.111027395003475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5761207342147827, "target_z": 0.4843750436500456, "ce": 5.728734970092773, "genuineness": 4.632867813110352, "identity": 0.5475167036056519, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d8f1644681b620e9b22ecad74743d2a06cfd4ae17fbe27f9762d165719fdcfc1.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "d94f12c2c4d422177aa5eb817727a6678ef3efdbb5ad3ad6f84e81ba6b7f124d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45abb465c177ee1ddf8827cc85f5a6e4a81c1a535cfae5c64def6955e97e5924", "generation_seconds_batch": 21.242282197927125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.401228427886963, "target_z": 2.0312499685719674, "ce": 6.046441555023193, "genuineness": 4.7834930419921875, "identity": 0.7110368013381958, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d94f12c2c4d422177aa5eb817727a6678ef3efdbb5ad3ad6f84e81ba6b7f124d.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d9a548e555640100f35c9eee070880688257fbbfab56ab0b4e237ba8316575c2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6abe7570889b091d489ee0179e52280d282f890d3d1e6837de71e8e3c4b5061a", "generation_seconds_batch": 23.698289455031045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16148324310779572, "target_z": -0.29296874170649134, "ce": 5.927301406860352, "genuineness": 4.406930446624756, "identity": 0.7108036875724792, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d9a548e555640100f35c9eee070880688257fbbfab56ab0b4e237ba8316575c2.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "d9b440602d8723c9f02193235a95ab2b19c8b0b28a30f1f83b1a2cd017a5a5ba", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a727fec96e2420ae9cc9d9983bd4e033fed5e178165c86b990b0170c5541a300", "generation_seconds_batch": 16.738844127976336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07188820838928223, "target_z": -0.46093749738099726, "ce": 6.125868320465088, "genuineness": 4.570107460021973, "identity": 0.6999863982200623, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d9b440602d8723c9f02193235a95ab2b19c8b0b28a30f1f83b1a2cd017a5a5ba.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "d9bb50065c3adc101655f91a1e408f5af4cbc69d6a4de12a3a1a1f760502591f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "08d9640a4a13c32b14227ad08c719206c41611f6ae16fcb9fc12c4b4f02abcc8", "generation_seconds_batch": 19.589918283978477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4875674843788147, "target_z": 0.3183593795832548, "ce": 5.974957466125488, "genuineness": 4.532451152801514, "identity": 0.5995916724205017, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur ja sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d9bb50065c3adc101655f91a1e408f5af4cbc69d6a4de12a3a1a1f760502591f.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "d9ed0ead8a6410a5934c14debd9825de38d640f86fd097918df764cf439c8f3a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "229687295eb76fb16a1945b0a0b743e1a5c106aa56831ecfd22d2f03f76b10d0", "generation_seconds_batch": 20.047504008980468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 6.173524379730225, "genuineness": 4.356721878051758, "identity": 0.39723217487335205, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/d9ed0ead8a6410a5934c14debd9825de38d640f86fd097918df764cf439c8f3a.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": "da4c78b3c3a8d2999ae8005df824ae211f3effa0dbaa5dea0757cdc5d90394c5", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96da0946dd931ca8e95f79e15e0113bb394508ec79891882db0971e136145d12", "generation_seconds_batch": 21.561631016083993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.976172924041748, "target_z": 1.234374959841958, "ce": 5.879645347595215, "genuineness": 4.168440818786621, "identity": 0.4548741579055786, "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/emo_Confusion/da4c78b3c3a8d2999ae8005df824ae211f3effa0dbaa5dea0757cdc5d90394c5.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": "dade76214f5964317fe17c1f5c546e5604404cade00541315ffebfadc5866372", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I wrote the number on the envelope.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9bbd389f58e3446e5a4e8b4446d8c7eac80b6a8f3cf74150aebe2fe44518fb3", "generation_seconds_batch": 16.487892873003148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5302814245223999, "target_z": 0.3984375602370629, "ce": 6.038498878479004, "genuineness": 4.344170093536377, "identity": 0.36435312032699585, "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/emo_Confusion/dade76214f5964317fe17c1f5c546e5604404cade00541315ffebfadc5866372.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": "daf150db51e01ebc9b1f9e1345df978898125f4e6ca9a2eced062767b080d561", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The keys are on the kitchen table.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1a4a9402179887dac84b4297dc55a274e996eb5bbabe4ebabb27402f30fc71d", "generation_seconds_batch": 21.561631016083993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0928548574447632, "target_z": 1.45312501920602, "ce": 5.935244083404541, "genuineness": 3.9927117824554443, "identity": 0.2970008850097656, "hypothesis": "The keys uh are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "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/emo_Confusion/daf150db51e01ebc9b1f9e1345df978898125f4e6ca9a2eced062767b080d561.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "2cb536e541cab7e3727541061562ddd3469861d0da24d1305ad5342e28f30193", "generation_seconds_batch": 15.203101939987391, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.403861999511719, "genuineness": 4.18099308013916, "identity": 0.2575709819793701, "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/emo_Confusion/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "db60ae34356dee3d5266a3e08008105589de40078cf4e62587bce7ebe0508ca0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "51829c6958cdee6fcc2d0b44f140afe98e4e5a64fa999453423568b2316ac4ca", "generation_seconds_batch": 13.980686991009861, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.015630871057510376, "target_z": -0.5664062321034813, "ce": 5.990842819213867, "genuineness": 4.482243061065674, "identity": 0.7655194401741028, "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/emo_Confusion/db60ae34356dee3d5266a3e08008105589de40078cf4e62587bce7ebe0508ca0.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "db9fbcfe90ebf20955540cd6d55154cbc1b5368e43b63745b8453cd1f70d6988", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c1908a6e2a0bdbf3a60f1cf0392909dee4a3d1cbbd5877d0d070e8c43951a8d5", "generation_seconds_batch": 20.73501576099079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4845726490020752, "target_z": 2.187499979047978, "ce": 5.816103935241699, "genuineness": 4.65797233581543, "identity": 0.7353223562240601, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/db9fbcfe90ebf20955540cd6d55154cbc1b5368e43b63745b8453cd1f70d6988.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "dba61081feaf3ddf94dbb1ae2f46579fb183838178a70a26b55bcdcbb7bfdcf9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "abba662a85b905ac73106f4982f738122b880d1671ff7857751a2a502d6cbd91", "generation_seconds_batch": 18.389347480027936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.126192569732666, "target_z": 1.515625068094071, "ce": 5.621508598327637, "genuineness": 4.406930446624756, "identity": 0.7322659492492676, "hypothesis": "Die ist noch offen. Ich hab 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/emo_Confusion/dba61081feaf3ddf94dbb1ae2f46579fb183838178a70a26b55bcdcbb7bfdcf9.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "dbbd59ed023e6654f6c28e764ab8ee669d16539ed2539a0ab3c3fcc881772220", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6fa3a371b2c3dc10020770b7d27551a852b2b66d2fe60a05601d23d3514419ac", "generation_seconds_batch": 20.91269912896678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6677993535995483, "target_z": 0.656250010476011, "ce": 5.871703147888184, "genuineness": 4.65797233581543, "identity": 0.7791010141372681, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/dbbd59ed023e6654f6c28e764ab8ee669d16539ed2539a0ab3c3fcc881772220.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "dc56ab7d6017104281e4ebdb5e9785985acd29a754bfb0a5671db843470c4e63", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6153bd69e45b95eed42aad80eebdcc84e40075845fa56107ba71b799308d217f", "generation_seconds_batch": 20.898207806050777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21565698087215424, "target_z": -0.1914062460714959, "ce": 5.816103935241699, "genuineness": 4.457139015197754, "identity": 0.647631824016571, "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/emo_Confusion/dc56ab7d6017104281e4ebdb5e9785985acd29a754bfb0a5671db843470c4e63.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "dc7f7e4691bf902bb17e5ffeaeb8943be319326db21ec1d9863f909fe029e407", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7a474d0aa14b92c74ddb4b2d99f6e034a50a357fdeab8bd57ea33ec40fae3df8", "generation_seconds_batch": 19.78072325210087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22294959425926208, "target_z": -0.17773438132925662, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.5730282068252563, "hypothesis": "Ich habe die kleine Schachtel gefunden. sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/dc7f7e4691bf902bb17e5ffeaeb8943be319326db21ec1d9863f909fe029e407.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "dd4876434582cb0237fecb3c0d2940e76e83a67396affde3be32b5aea51f1cce", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd175830e44e04a18a4576bc9db8cfa4087c035bfc2c1d763a71598d6a4a35dc", "generation_seconds_batch": 21.233912243973464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07188820838928223, "target_z": -0.46093749738099726, "ce": 6.38003396987915, "genuineness": 4.595211505889893, "identity": 0.16822627186775208, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/dd4876434582cb0237fecb3c0d2940e76e83a67396affde3be32b5aea51f1cce.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "dd817c043913bba8367b447e78e4616de362edcbe38f6cac943c779c51902b84", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6c49d464ff6f24e6dcb7e1fd1124709660dd9324cc71aa42732e68b8673657a5", "generation_seconds_batch": 15.999598938971758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0428483486175537, "target_z": 1.35937505761806, "ce": 5.903473377227783, "genuineness": 4.7081804275512695, "identity": 0.6632610559463501, "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/emo_Confusion/dd817c043913bba8367b447e78e4616de362edcbe38f6cac943c779c51902b84.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "ddc86ccd2991981e2404caf764a97fcf80e058ae876eeea866f6a4c73167f233", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ee0b9ac829b1b3cd3df9b91f6bba5e3dfe8702334b833c353088ad9715d385a5", "generation_seconds_batch": 20.692741798004135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 6.070269584655762, "genuineness": 4.595211505889893, "identity": 0.7377275228500366, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ddc86ccd2991981e2404caf764a97fcf80e058ae876eeea866f6a4c73167f233.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "dde0c95db385eae5e00a53b6aa6cf8a5f7b27024936abb28921916ec71e67270", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "abc54c2e304018baa44adfdffab3d525e565d495ae1cbe92c9814371b183168c", "generation_seconds_batch": 21.17464775289409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6012545824050903, "target_z": 2.40625003841204, "ce": 5.772419452667236, "genuineness": 4.444586753845215, "identity": 0.6509719491004944, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/dde0c95db385eae5e00a53b6aa6cf8a5f7b27024936abb28921916ec71e67270.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "deeada477e66c8b524145b49b22ea0fc5aacfd19212ed4a112dc266c4431bd50", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81afe8800277fd85be3e01fd064ad18bba84cb08598b313e1253f99f73bf47f1", "generation_seconds_batch": 18.237061318941414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2953548729419708, "target_z": -0.04199220987064836, "ce": 5.935244083404541, "genuineness": 4.319066047668457, "identity": 0.34717705845832825, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/deeada477e66c8b524145b49b22ea0fc5aacfd19212ed4a112dc266c4431bd50.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "df61a9b1b55cfe668cbc1d021a970e2083c98c6d4a76ae83257ccf34e5576a29", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f979392660834c392f09a048fa49d0cd313b090650ba1f116af33aea794506f3", "generation_seconds_batch": 21.37910819402896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5762512683868408, "target_z": 2.3593749458739435, "ce": 5.585766792297363, "genuineness": 4.444586753845215, "identity": 0.638109564781189, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/df61a9b1b55cfe668cbc1d021a970e2083c98c6d4a76ae83257ccf34e5576a29.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "df8ee8142ee15a3821035c901b042e55aad0e4080719835572f0c554a2f5ee8f", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a2da1965b2f755ac2d35c30eff7a4837a5c64caff1bf81ac7fcfeebcbfbdb994", "generation_seconds_batch": 13.993220232077874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 5.919358730316162, "genuineness": 4.607763767242432, "identity": 0.49458932876586914, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/df8ee8142ee15a3821035c901b042e55aad0e4080719835572f0c554a2f5ee8f.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e0662497e10a8c9ac9df0d6a8e3449245f8dca8086ce20993c47e80ad97184ce", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3f23e35738adbd50442a93c20ed1e55e202e113c27828b740b760a1d7bb6364", "generation_seconds_batch": 14.121279977960512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9970090389251709, "target_z": 1.2734375742050774, "ce": 5.831989288330078, "genuineness": 4.3127899169921875, "identity": 0.3349449634552002, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e0662497e10a8c9ac9df0d6a8e3449245f8dca8086ce20993c47e80ad97184ce.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "e0db82480be3fd57036472890b155ac5a25a8e2750c5be706dcf25a9c3e52c8b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "531fce0b625c09c84b9dc25582e927fb55404babd75952955ff6a852d2b1afd4", "generation_seconds_batch": 15.93824567401316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6448796987533569, "target_z": 0.6132812687695196, "ce": 5.903473377227783, "genuineness": 4.507347106933594, "identity": 0.7244918942451477, "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/emo_Confusion/e0db82480be3fd57036472890b155ac5a25a8e2750c5be706dcf25a9c3e52c8b.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "e0df356506fd5f5b5a45d9e60c652555aae0f868a1a9068a1d6479974836df0d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fcd1ccb764b7d9768fc7a849847affc79643b9ef75e756eea7409ede859e36ac", "generation_seconds_batch": 22.087910200934857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017714470624923706, "target_z": -0.5624999930159927, "ce": 5.863759994506836, "genuineness": 4.833701133728027, "identity": 0.6523470878601074, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e0df356506fd5f5b5a45d9e60c652555aae0f868a1a9068a1d6479974836df0d.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "e2707d18100d682c19c95e3e250b1623f05bd20d8ba5719a8c2a750190e196df", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "785eb52167c80783d0883b20acf884c4a3dbf4cc3ff3eb0db379f0a5d860eeb8", "generation_seconds_batch": 20.768599046976306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1013197898864746, "target_z": 3.343749877779872, "ce": 5.649308204650879, "genuineness": 4.193545341491699, "identity": 0.2946184277534485, "hypothesis": "De deur staat toch 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/emo_Confusion/e2707d18100d682c19c95e3e250b1623f05bd20d8ba5719a8c2a750190e196df.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": "e29393669cb584cb6a34360aa5f3efb69a9ab0fb1b7255f11088f379b78117ca", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e1df6c9148a547712b3472e92fa793a7c0639d71a51e372031605705149fb92", "generation_seconds_batch": 20.982111165998504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8761599063873291, "target_z": 1.0468750366660382, "ce": 5.863759994506836, "genuineness": 4.6705241203308105, "identity": 0.7817909717559814, "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/emo_Confusion/e29393669cb584cb6a34360aa5f3efb69a9ab0fb1b7255f11088f379b78117ca.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "e2cc319aa2ea9f357955dfb03d997f0c2dc65385e8d594c8ae4945b8b8cbff24", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "92b74cb053e9c00c8fbefaf1d7b41b423d133bb52414c3d1c1e9d5a1c22e417e", "generation_seconds_batch": 24.39151219499763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04271772503852844, "target_z": -0.5156250122220127, "ce": 6.41180419921875, "genuineness": 4.595211505889893, "identity": 0.32501593232154846, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e2cc319aa2ea9f357955dfb03d997f0c2dc65385e8d594c8ae4945b8b8cbff24.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "e355e719b7054e9c5eba4c0e097cba30a4ff664e1d69950fb96662b0996c16f7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3bdc179a132b2327c1f594a3cbc16155860ade3025153d53372c194a89fc379", "generation_seconds_batch": 21.4163426309824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.659464955329895, "target_z": 0.6406250541260565, "ce": 5.454712867736816, "genuineness": 4.331617832183838, "identity": 0.5518674850463867, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e355e719b7054e9c5eba4c0e097cba30a4ff664e1d69950fb96662b0996c16f7.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "e3e09e8cf5cb8a64fa7110721494002f64ab16848ea96b09f37baf170b4c18d9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1e64ae2a2e94555ed0472b90d2f94329ace531e7f083cbef18786b6129da42b2", "generation_seconds_batch": 20.193781876005232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07918082177639008, "target_z": -0.44726563263875796, "ce": 5.982900142669678, "genuineness": 4.532451152801514, "identity": 0.7552136182785034, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e3e09e8cf5cb8a64fa7110721494002f64ab16848ea96b09f37baf170b4c18d9.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "e404b495d4556c0d8f36200efa0e719a3660baca62625d0bc0a7b67a95661fb5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "34c3ae72e7a58714ef8ed5599986131e00422e32f2626786a39fa9016fefb341", "generation_seconds_batch": 20.532968975952826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.268008232116699, "target_z": 3.656249898731894, "ce": 5.007937431335449, "genuineness": 4.400654315948486, "identity": 0.5541314482688904, "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/emo_Confusion/e404b495d4556c0d8f36200efa0e719a3660baca62625d0bc0a7b67a95661fb5.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e451fa910a15cd96c3b643e74ab2a65a93a70ba45f517c1b133689c21a2d2d00", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ad88e9ba0f7824a22ba5fd09fac1bd9e96dafdafab92e12438d3a6867190f0f", "generation_seconds_batch": 23.829392445040867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 6.149696350097656, "genuineness": 4.193545341491699, "identity": 0.3335621953010559, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e451fa910a15cd96c3b643e74ab2a65a93a70ba45f517c1b133689c21a2d2d00.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": "e45cc0302769f26f52cecd94ae09aa7076a90458548aeac603370050bffdbe39", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a248f734ac2b630642ffed6c8522c958a3b65d6616a3c18d3a641a0a9439b606", "generation_seconds_batch": 20.381754116970114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.359686851501465, "target_z": 3.8281248655578595, "ce": 5.756534099578857, "genuineness": 4.306513786315918, "identity": 0.6311932802200317, "hypothesis": "Die Schlüsse 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/emo_Confusion/e45cc0302769f26f52cecd94ae09aa7076a90458548aeac603370050bffdbe39.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": "e477df3b7f8467023752b716fb0c57aadc05c13b726fcaf410ec440b3ae3bc13", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d3e6ceaf0eae4335f17a22d43ef51fb4870bf478b55d86438fec451d27d27773", "generation_seconds_batch": 21.457923726993613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9012937545776367, "target_z": 2.9687500314280326, "ce": 5.411027908325195, "genuineness": 4.256305694580078, "identity": 0.24337278306484222, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e477df3b7f8467023752b716fb0c57aadc05c13b726fcaf410ec440b3ae3bc13.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "e5889b0c28623c9521e07eceb1342195afb5d2d59191220c677638d2897c86d0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "935775c65c85af5bff589dbaf43c5a48c259a0fbbf84c7487d06463677f2e139", "generation_seconds_batch": 21.789721698034555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1595301628112793, "target_z": 1.5781248934938887, "ce": 5.137005805969238, "genuineness": 4.400654315948486, "identity": 0.5964210629463196, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e5889b0c28623c9521e07eceb1342195afb5d2d59191220c677638d2897c86d0.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "e6019629587affc9bf5b09432f58eba75d1b5440df1f3f4f0bfbd612c3473ea9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "517501ce0d8f06dd3d3430b43edb891643975d77868b902a9178aee126d674f2", "generation_seconds_batch": 20.880221002968028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0812644362449646, "target_z": -0.4433593656152402, "ce": 5.8875885009765625, "genuineness": 4.607763767242432, "identity": 0.5048834681510925, "hypothesis": "Ma 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/emo_Confusion/e6019629587affc9bf5b09432f58eba75d1b5440df1f3f4f0bfbd612c3473ea9.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "e695fb1d11a0b8d75e5741f5bfdc768f3800d0a2ce32d931b2ab01de24ff2538", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8fc7f22da0303597653f3011c7eb5132026e9d3e7efb6cc1385e91536da53465", "generation_seconds_batch": 26.950108854100108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.292664527893066, "genuineness": 4.557555198669434, "identity": 0.43537694215774536, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e695fb1d11a0b8d75e5741f5bfdc768f3800d0a2ce32d931b2ab01de24ff2538.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "e69e2fe910a289386a6be63325d12f1fb71fdcb635fe71de7d632024c1463ea9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5bb0c983c37719051acb3c49581ee894213952e531769dac260566d79fd317f9", "generation_seconds_batch": 15.781815703958273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3720580339431763, "target_z": 1.9765626213471268, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.5979759693145752, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e69e2fe910a289386a6be63325d12f1fb71fdcb635fe71de7d632024c1463ea9.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e6a40163ad31136b4777001cb700a311f6b2c0bdfa2a09883e36de017a87458f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81115b1618a91a1915347a273aa8882ccfd4a137859dbaef5b804e6b5916bfdd", "generation_seconds_batch": 22.31962552596815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.052093952894210815, "target_z": -0.4980468804562557, "ce": 6.205295085906982, "genuineness": 4.325342178344727, "identity": 0.37009525299072266, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e6a40163ad31136b4777001cb700a311f6b2c0bdfa2a09883e36de017a87458f.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": "e6b7da7a5f3dc8fa075540cd30322facbcaf92e1cdfe7fbba55f65d78fec82a6", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7489bee19b664572c322bc66d0a425f811b4b87cd5c1bb2d0c32bf15058a93d4", "generation_seconds_batch": 21.139035118976608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.667930006980896, "target_z": 2.531250136188142, "ce": 5.450741291046143, "genuineness": 4.369274139404297, "identity": 0.6349796056747437, "hypothesis": "Ihr Mann hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "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/emo_Confusion/e6b7da7a5f3dc8fa075540cd30322facbcaf92e1cdfe7fbba55f65d78fec82a6.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "e6bd3a55cbcc29f4b469a55b5c7dc3b9f18d0954c9bb23a087477ab700ab8168", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 130, "seconds_generated": 10.4, "mp3_sha256": "9364324ed3f8e61054dcb45f0782522a3e206fad549f6744e40db3228f335d2d", "generation_seconds_batch": 21.71078587102238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2364930361509323, "target_z": -0.15234374345249316, "ce": 5.359400749206543, "genuineness": 4.519899368286133, "identity": 0.39139434695243835, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e6bd3a55cbcc29f4b469a55b5c7dc3b9f18d0954c9bb23a087477ab700ab8168.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "32f1ab57c420c52d94f694a0dd643f89391e1fe67f6601aedb9973ee9611c597", "generation_seconds_batch": 15.965634776977822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07605542242527008, "target_z": -0.4531249912699909, "ce": 6.252951145172119, "genuineness": 4.375550270080566, "identity": 0.48762261867523193, "hypothesis": "Someone moved the little blue chair it used to stand here beside the t tall bookshelf. We can put it back after lunch.", "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/emo_Confusion/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e6f6eabba835b67bac2156816e49907aea32ac764bd53bad1624d737942498bc", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f866b42d07b6dcab08552652a0b37fae372218fe951fd9b46ff331b98d3848f2", "generation_seconds_batch": 21.352340943994932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7261402606964111, "target_z": 0.7656249284139253, "ce": 5.569881439208984, "genuineness": 4.111956596374512, "identity": 0.24635308980941772, "hypothesis": "The door is still open. I I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e6f6eabba835b67bac2156816e49907aea32ac764bd53bad1624d737942498bc.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "e760fdd289bff7c06c9e2329b998d1c26102aaab65d4789c69b72501e128d3da", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9735cbde8bb4e0a1d49d6003408ebc28fc33e5f3561f5f40195941a4ba7a228d", "generation_seconds_batch": 24.631402708939277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05313575267791748, "target_z": -0.4960937609125114, "ce": 6.157639026641846, "genuineness": 4.645420074462891, "identity": 0.6867939233779907, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e760fdd289bff7c06c9e2329b998d1c26102aaab65d4789c69b72501e128d3da.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "e7624956ae87456933002583caeb7a3e3d571dcc6601e8c8a8236978ad8c0610", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "960ece4460ed0a8893617a88e14a7bbcacba026b1244648e23e1807bacb4b688", "generation_seconds_batch": 16.503352447994985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8386549949645996, "target_z": 0.97656250960301, "ce": 5.34748649597168, "genuineness": 4.482243061065674, "identity": 0.6563615798950195, "hypothesis": "j'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e7624956ae87456933002583caeb7a3e3d571dcc6601e8c8a8236978ad8c0610.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e767818f60bb69ba076c6f94556b25e6243191f5f3ce849bab992c2369ddd45d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8311f068d5fdac18dfde83463a98e837d23b7a472f7af08f57a7b3e211fe2247", "generation_seconds_batch": 21.468618754995987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.198988139629364, "target_z": -0.22265624257949226, "ce": 5.677107334136963, "genuineness": 4.068024158477783, "identity": 0.4017043113708496, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e767818f60bb69ba076c6f94556b25e6243191f5f3ce849bab992c2369ddd45d.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "e7e2112661e6676ff0b77e459ac792f8a6019dd91a516beb187da7bb197e1c0a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1bfe2fc706f6b64472ffb2fbea8125934668a3b3ecd6f879a488f49c96071133", "generation_seconds_batch": 19.06448508996982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9346314668655396, "target_z": 3.031250080316084, "ce": 5.863759994506836, "genuineness": 4.250029563903809, "identity": 0.45082634687423706, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e7e2112661e6676ff0b77e459ac792f8a6019dd91a516beb187da7bb197e1c0a.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "e84b90f0c6c0ca829614a7bc97e430d9287e6d66c66b2189ee7492652044dea6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "85894d6200b2e727d168540719140e0389c31e352a178b102e46adbbdb7cb90f", "generation_seconds_batch": 22.686378529062495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.842952847480774, "target_z": 2.8593751134901186, "ce": 5.071478843688965, "genuineness": 4.13078498840332, "identity": 0.6589688062667847, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e84b90f0c6c0ca829614a7bc97e430d9287e6d66c66b2189ee7492652044dea6.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "e8e80e791be8d20aa585605700f569d564343c09323e5d3e2dba0613abcbf869", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6e44b9e614241b133d5007a68855c10ea461b62afe26fa607d27426b0ec93078", "generation_seconds_batch": 27.409488159930333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7948992848396301, "target_z": 0.8945312652775159, "ce": 5.728734970092773, "genuineness": 4.821148872375488, "identity": 0.6924716234207153, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e8e80e791be8d20aa585605700f569d564343c09323e5d3e2dba0613abcbf869.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e8eaa15ad04b92672808304bd872469788a4e7cad25e3264793789a0c7802076", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a57bfb04eca762dcbd88413ec7142c360dc08d4d7897cce2b020d43b89fe84a", "generation_seconds_batch": 24.316069032996893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 5.824047088623047, "genuineness": 4.093128204345703, "identity": 0.39249181747436523, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e8eaa15ad04b92672808304bd872469788a4e7cad25e3264793789a0c7802076.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "e94f71672843b263e2d3b6c3808e9b517dbdc7139a33b44b77d7287dae7a612a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1010dbfcd596ecc664e714b509a0b546f34bc012a7cae4a23a25149a274d9543", "generation_seconds_batch": 20.075562836020254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05313575267791748, "target_z": -0.4960937609125114, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.33158886432647705, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e94f71672843b263e2d3b6c3808e9b517dbdc7139a33b44b77d7287dae7a612a.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "e9ce961f89a5da64554322bd56e81f0b626dea7d558b2e2c3c572466e1746cc7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5fd728f6044fe1784897a2f02ce84084ddce2794f7802ce2c51ef27d9283f531", "generation_seconds_batch": 19.668697881978005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0011762380599976, "target_z": 1.2812500523800547, "ce": 5.967014789581299, "genuineness": 4.344170093536377, "identity": 0.4483943581581116, "hypothesis": "I found the little box. It was beside the old window we can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/e9ce961f89a5da64554322bd56e81f0b626dea7d558b2e2c3c572466e1746cc7.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "80769ee22c1e67fdf72530e7b93a9beb5b6bec98aad150aec760e1b720309179", "generation_seconds_batch": 14.662117762025446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05105215311050415, "target_z": -0.5, "ce": 6.427689552307129, "genuineness": 4.406930446624756, "identity": 0.4904756546020508, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem Hohenbücherregal. Wir stellen ihn nach dem Essen zurück.", "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/emo_Confusion/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "e9e514491d23b1e4bd0656d4b509a7d70803fdd623f8f72dc9a276e674707d1e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "810bac584b8a290d199eabe09fd062eec228dcea8d93e50c3ec583f3b32bb9ae", "generation_seconds_batch": 19.679027174017392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5345792770385742, "target_z": 2.2812501641241716, "ce": 5.3514580726623535, "genuineness": 4.137060642242432, "identity": 0.5928807258605957, "hypothesis": "J'ai trouvé la petite boîte. Elle était. 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/emo_Confusion/e9e514491d23b1e4bd0656d4b509a7d70803fdd623f8f72dc9a276e674707d1e.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "ea9725d7bf5f79666f9a1e1a3d8a46111737754a410be0ce17e0f4a7a02eb156", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a29ae29ff1f5fc732ff126e98b5e9429b296a7d72973f333db87171bd4a4f27f", "generation_seconds_batch": 20.169644415960647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9262970685958862, "target_z": 3.0156251239661294, "ce": 6.09409761428833, "genuineness": 4.494794845581055, "identity": 0.4771616458892822, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ea9725d7bf5f79666f9a1e1a3d8a46111737754a410be0ce17e0f4a7a02eb156.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "ea97d3bc572f892b58e59d2b7cf57cbcc4bb7c1507f9bb871e8cf53eddc4ec8a", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8f786f8b1b299624a9b28e7a4651d098dbbc611482655926696441285d386c05", "generation_seconds_batch": 15.536983537953347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05730298161506653, "target_z": -0.4882812268654758, "ce": 6.006728172302246, "genuineness": 4.645420074462891, "identity": 0.5092672109603882, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ea97d3bc572f892b58e59d2b7cf57cbcc4bb7c1507f9bb871e8cf53eddc4ec8a.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "eb6d932c9f0fd15cfebded2c696612041bc3284d4074b67c17da6f101d8ba829", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f74a41c9de6baf4cab14c4a4f5523e3f545e456283a3014e3500fe7278122ba5", "generation_seconds_batch": 20.67536067601759, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.062326908111572, "genuineness": 4.7458367347717285, "identity": 0.8345397710800171, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/eb6d932c9f0fd15cfebded2c696612041bc3284d4074b67c17da6f101d8ba829.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "eb739e5ea4c8f20a150d97f39cfc601ffab74f03e4806a3b908a70c9c7343ec2", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "06aaab2ac95c9808416779b12b308808d6b5969a6069475f539ab3802761bee8", "generation_seconds_batch": 20.736993049038574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14898160099983215, "target_z": -0.3164062600395105, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.5477709174156189, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/eb739e5ea4c8f20a150d97f39cfc601ffab74f03e4806a3b908a70c9c7343ec2.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "1870065f6e88784714ac4c24d2e049cd1f19024c8e2a561c98f35c6ca3aab01a", "generation_seconds_batch": 15.385275140986778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05938658118247986, "target_z": -0.4843749877779872, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.7223421931266785, "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/emo_Confusion/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "ec0dbc5e5cb3b1dd153ee38d4e715c2817d6f8001f49a46e1023475e17698477", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21998cba4f53174467c7e5abb47570261c5b0de47a98b1a0069f6e60b3369968", "generation_seconds_batch": 13.565647377050482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8929593563079834, "target_z": 2.9531250750780784, "ce": 5.657250881195068, "genuineness": 4.193545341491699, "identity": 0.7898534536361694, "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/emo_Confusion/ec0dbc5e5cb3b1dd153ee38d4e715c2817d6f8001f49a46e1023475e17698477.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "ec6ef6e85efe22ca6afca8f41893ab0a232267d687d71df89412150e36040715", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "29245e3ccd04ec4cbf55e1707e4b205af0f51e54fe258fd6428d78c83ea1665f", "generation_seconds_batch": 14.839654676034115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00104561448097229, "target_z": -0.5937500174600182, "ce": 6.300607204437256, "genuineness": 4.2374773025512695, "identity": 0.6342116594314575, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ec6ef6e85efe22ca6afca8f41893ab0a232267d687d71df89412150e36040715.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "ec90c86194f4cf33c7ce7f3e17b868e27c05b3c4e51b5273cdfd1a5755c6405f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4d47f57154bfebfb79c11d558e192364066399f7bbdb5b4f34e4e457d555d2ef", "generation_seconds_batch": 20.721714751096442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9428352117538452, "target_z": 1.171874910953907, "ce": 6.24500846862793, "genuineness": 4.68307638168335, "identity": 0.5641500949859619, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ec90c86194f4cf33c7ce7f3e17b868e27c05b3c4e51b5273cdfd1a5755c6405f.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "edd55cbc5d611e1995b04894d83d0223114c8f8e99c11a50d5c285cb654ac5d5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e63e7647cea47157dea138709463840f8f791d196947b6be6451a63c0ed501cb", "generation_seconds_batch": 21.201141183031723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.988674521446228, "target_z": 1.2578123943668897, "ce": 5.871703147888184, "genuineness": 4.3127899169921875, "identity": 0.3385685086250305, "hypothesis": "the door i is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/edd55cbc5d611e1995b04894d83d0223114c8f8e99c11a50d5c285cb654ac5d5.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "eea08d9dca22fdea97d1d1c8155d92628bc63cf233273aedc60e07ca8a973091", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d058a9544474344b029de092abf434927e5e8c67cddcaf0fbc272dd221aed03b", "generation_seconds_batch": 16.276890425011516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.2767791748046875, "genuineness": 4.080576419830322, "identity": 0.4608805179595947, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/eea08d9dca22fdea97d1d1c8155d92628bc63cf233273aedc60e07ca8a973091.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "eea8fb10e9fb59da93b423922021046b99ea9d6270b8d336c3b761c220ee87ed", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47647b52d9ed0fb2b3435e46a4f158152e1b07ed5176b0ff1d36629f4f5dec95", "generation_seconds_batch": 25.46120472298935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8469893932342529, "target_z": 0.9921874659529645, "ce": 5.569881439208984, "genuineness": 4.570107460021973, "identity": 0.6604534387588501, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/eea8fb10e9fb59da93b423922021046b99ea9d6270b8d336c3b761c220ee87ed.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "ef6f8aa18a3bc91fa7efeab4c0c8daf1ceba544515f11bd288978bfe9213ca04", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ee888c4717d3bdb2b1509ee61d1dfd152f6881c919e3434783be51f96b903d6a", "generation_seconds_batch": 22.087910200934857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04271772503852844, "target_z": -0.5156250122220127, "ce": 5.569881439208984, "genuineness": 4.65797233581543, "identity": 0.7459824085235596, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ef6f8aa18a3bc91fa7efeab4c0c8daf1ceba544515f11bd288978bfe9213ca04.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "efae0b24a4a08f81500e3c627ee146eb76e9f09106a8a1bcc08a994517dec9a5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ff099808beb3ce351a3d81c73cb88b234e6d52b1b81d4d895bd2e8422a2db894", "generation_seconds_batch": 21.928549460950308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8594909906387329, "target_z": 1.015624900477896, "ce": 5.935244083404541, "genuineness": 4.432034492492676, "identity": 0.45660504698753357, "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/emo_Confusion/efae0b24a4a08f81500e3c627ee146eb76e9f09106a8a1bcc08a994517dec9a5.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "f01d9cd8f66ea620817ed74d3d904f1408477d3661dc0501d61c8c9e9a9cd9c8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2449a1d87be54b8b028b97362ad6c3b34f8bc0a197ff38de8ecba92045668ad", "generation_seconds_batch": 19.705652939970605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09168246388435364, "target_z": -0.42382811430573886, "ce": 5.879645347595215, "genuineness": 4.432034492492676, "identity": 0.21938851475715637, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f01d9cd8f66ea620817ed74d3d904f1408477d3661dc0501d61c8c9e9a9cd9c8.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "f02b5326fae9604153ad5553ea35cf3fbabf436582090d4dfd63afc88f4023aa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "11ed4a92859186e02cfb74e62ccb00901e1f810b6b116c22dd449315de51e5cc", "generation_seconds_batch": 24.39151219499763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.221180438995361, "genuineness": 4.287685871124268, "identity": 0.47196388244628906, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f02b5326fae9604153ad5553ea35cf3fbabf436582090d4dfd63afc88f4023aa.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "f0814af4b56ec286d31c54e4c6ddea015810dbfc2b77e30f0ed6b84f33978019", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e32af3128e6caae4bec822b0f34ae8db68ab3efcca8b7144e5a2da335a9a0faf", "generation_seconds_batch": 19.655113793094642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19430002570152283, "target_z": -0.2314453084623708, "ce": 5.919358730316162, "genuineness": 4.8588056564331055, "identity": 0.7397983074188232, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f0814af4b56ec286d31c54e4c6ddea015810dbfc2b77e30f0ed6b84f33978019.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "f09e82e4fda511aa6bb7116b520f6b8b8d20b53c3d0dbd488718a239b03d31ed", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "66737fa543dcbdd9b357ae08ced8460a07f82f4f507a76e6ac47a46b1fc449ee", "generation_seconds_batch": 20.135928080999292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42141300439834595, "target_z": 0.19433592538711236, "ce": 5.681078910827637, "genuineness": 4.6705241203308105, "identity": 0.6845471858978271, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f09e82e4fda511aa6bb7116b520f6b8b8d20b53c3d0dbd488718a239b03d31ed.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": "f0acfe1f3a522f2691019b0ce417c21523108472534bee7c4d81d5b213515a27", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd4052731d990fc9ea8211fc9c73eb3ca25d2ea24a8f7aee76050bed30317b82", "generation_seconds_batch": 15.319388268049806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4095629453659058, "target_z": 2.046875148410155, "ce": 5.4666266441345215, "genuineness": 4.632867813110352, "identity": 0.5919102430343628, "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/emo_Confusion/f0acfe1f3a522f2691019b0ce417c21523108472534bee7c4d81d5b213515a27.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "f1b2e84807159aff5792759e035f4118cfcd6ffd7ef89b4d13f960b255db3a6c", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7df8158290b8e42772416cc4c3bd374400142bf6ae735a89737b0036e99c89df", "generation_seconds_batch": 20.374852536944672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4095629453659058, "target_z": 2.046875148410155, "ce": 5.959072113037109, "genuineness": 4.268857479095459, "identity": 0.5923872590065002, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f1b2e84807159aff5792759e035f4118cfcd6ffd7ef89b4d13f960b255db3a6c.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "66e241ef2028bb3f23f0c9e6771bded8cd4da80c3de912a28b517383a5f9dd3c", "generation_seconds_batch": 15.293119743932039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18961191177368164, "target_z": -0.24023437434524933, "ce": 6.23706579208374, "genuineness": 4.582659721374512, "identity": 0.6910632848739624, "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/emo_Confusion/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "f2e521ed9ddc82c5dcff9775bd746e8697d6dbe4483f344a91b8765c669e37a6", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e50fc0e4b92ffdfd245dd097fffbe23bd898b4949f084ec4bf441d3184e4b09d", "generation_seconds_batch": 16.17279724299442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2637104988098145, "target_z": 1.773437518333019, "ce": 6.006728172302246, "genuineness": 4.375550270080566, "identity": 0.6621191501617432, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd: Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f2e521ed9ddc82c5dcff9775bd746e8697d6dbe4483f344a91b8765c669e37a6.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": "f2f3a1c387dc375c88db856573c4a0ad38b57cad1460a811da0711e93c5bbaa9", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bedb6dfb69fce24f1e41c820d6cf7ec67fb13e148230e5480929e4dcd4cd2b3", "generation_seconds_batch": 13.565647377050482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3595563173294067, "target_z": 1.9531249633339618, "ce": 5.577824115753174, "genuineness": 4.607763767242432, "identity": 0.45770198106765747, "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/emo_Confusion/f2f3a1c387dc375c88db856573c4a0ad38b57cad1460a811da0711e93c5bbaa9.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "f2f91f3f40b14352f96e17d01999f7e688b05558e5d68adb40d2e52fbaee7143", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "fbd5c8faa1081c8ed1e30b18b60c8268d10ae41e9a056e90bf7521e9926f01c2", "generation_seconds_batch": 20.714682637015358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.774063229560852, "target_z": 0.8554687626585132, "ce": 5.959072113037109, "genuineness": 4.457139015197754, "identity": 0.28503483533859253, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f2f91f3f40b14352f96e17d01999f7e688b05558e5d68adb40d2e52fbaee7143.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "f2ff9ed0988cb275ee9f8ce2bfd461f32e02af570b42eb67a725fe830f4a6edf", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "154a7bf79d4e694c93d764eab214a685f2ae2d42a64501cad04a4ffd3191d64f", "generation_seconds_batch": 20.808632764965296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.202113538980484, "target_z": -0.21679688394825936, "ce": 5.561938762664795, "genuineness": 4.758388519287109, "identity": 0.7408088445663452, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f2ff9ed0988cb275ee9f8ce2bfd461f32e02af570b42eb67a725fe830f4a6edf.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "f327c7dd831c68d33760df86e22377b3edbf12f1ee8e30614b5b992bc61bc286", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3ef0d1bcd360e1af6410c3d83e88555f44bae25ee7089a1264c1373259329376", "generation_seconds_batch": 21.459266362944618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9470024108886719, "target_z": 1.1796873891288842, "ce": 6.006728172302246, "genuineness": 4.218649387359619, "identity": 0.6670984029769897, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f327c7dd831c68d33760df86e22377b3edbf12f1ee8e30614b5b992bc61bc286.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "f3b654fbefd8a8b41d4e96971c72c98018af493f2977b4da286e78c52ad622b8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "67b6eef12de1140ba69a8c4e26af57764d6cd79018638c0d751dd065d2445058", "generation_seconds_batch": 20.763286706991494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34002217650413513, "target_z": 0.041748048975658446, "ce": 5.450741291046143, "genuineness": 4.319066047668457, "identity": 0.6228517293930054, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f3b654fbefd8a8b41d4e96971c72c98018af493f2977b4da286e78c52ad622b8.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "f4b296814e6747e61e4919041f3496fe0881d8a5a20a3a5b932351484a6b8bba", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1408d5cd916c0874d2d61dcab84e9942835875804d8d76fb3148ca14b7a793b0", "generation_seconds_batch": 29.028887607040815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3220515251159668, "target_z": 1.8828126597591668, "ce": 5.9431867599487305, "genuineness": 4.369274139404297, "identity": 0.7007580399513245, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f4b296814e6747e61e4919041f3496fe0881d8a5a20a3a5b932351484a6b8bba.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "f55aece7b17cffca674bf5bb77c87174fee5b5e72b47b69dceb00e9e1dca72fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "43997e03652f69812dd73894b7866fe8d45b27ad308569038a40c32ad9c7d7a2", "generation_seconds_batch": 25.187727207085118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07605542242527008, "target_z": -0.4531249912699909, "ce": 6.006728172302246, "genuineness": 4.494794845581055, "identity": 0.7126359939575195, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f55aece7b17cffca674bf5bb77c87174fee5b5e72b47b69dceb00e9e1dca72fe.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "f59a4cd7df6ddd9b51c6b20fe818ebf164381a9d06222f06bae630fd234ccb77", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ec3990df82ae3aec3a75701a3587eb774637be1ee4fb0afcd5d2d9deac897259", "generation_seconds_batch": 21.459266362944618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3763556480407715, "target_z": 3.8593747782577683, "ce": 5.974957466125488, "genuineness": 4.432034492492676, "identity": 0.3151199519634247, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar 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/emo_Confusion/f59a4cd7df6ddd9b51c6b20fe818ebf164381a9d06222f06bae630fd234ccb77.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "f5faef1424258e49659ef99c0dfff128cf8be65e3a949ccb4a1cbb8d70a4f016", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1d0ac008a5fae7c32b90e83252a1198abdc5ffa07cd9fd67cb11b1af269969bf", "generation_seconds_batch": 27.377952252049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2928810119628906, "target_z": 1.828125089046093, "ce": 5.998785495758057, "genuineness": 4.444586753845215, "identity": 0.3312002420425415, "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/emo_Confusion/f5faef1424258e49659ef99c0dfff128cf8be65e3a949ccb4a1cbb8d70a4f016.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "f656f8c442db4a6154353a3f4c28cffa641c047c76a83380ee3f733830a02091", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e7e2e9044dac893ddfb383d31092ea8a92daa2c2953207b23272f4478fc143f3", "generation_seconds_batch": 14.931639419985004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.6568666100502014, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f656f8c442db4a6154353a3f4c28cffa641c047c76a83380ee3f733830a02091.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "f6e69e981ec9b2ed28a77676e178bb18a6e6036b7d8666341c925e6a68e00f18", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5ffdbfad22125d7d6e62658723540f086848ba00b35d471326c97e8e3c3c8772", "generation_seconds_batch": 20.692741798004135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.013547241687774658, "target_z": -0.5703125270630283, "ce": 6.0305562019348145, "genuineness": 4.557555198669434, "identity": 0.6201009154319763, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f6e69e981ec9b2ed28a77676e178bb18a6e6036b7d8666341c925e6a68e00f18.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "510bbd0c06efa4b582a3be94a6cfbf5a342fe8c927169cf6861e4e8b29252250", "generation_seconds_batch": 15.621403019991703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 6.157639026641846, "genuineness": 4.68307638168335, "identity": 0.7984129190444946, "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/emo_Confusion/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "f805294e0fb5c5ec2dca1ea925114574b6adbcf5fb0d7cf7ebe672613b4fc174", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "48c1a9bf6abc67aa04567a062cc0478bd38b3156ef9591d977b5d84bbb959e8f", "generation_seconds_batch": 21.117626238963567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6907190084457397, "target_z": 0.6992187521825023, "ce": 5.7247633934021, "genuineness": 4.319066047668457, "identity": 0.5987064242362976, "hypothesis": "Ik heb een klein doosje 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/emo_Confusion/f805294e0fb5c5ec2dca1ea925114574b6adbcf5fb0d7cf7ebe672613b4fc174.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": "emo_Confusion", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": "29468882939771294ce5defb2621983d23e61d5ede8d3dcd1eee0d3402474a39", "generation_seconds_batch": 14.294577624998055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.6362618803977966, "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/emo_Confusion/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "f87293a6cea3b4ed5b67bf6614f8daa46479a597c55c442a8b79bce4b14f5658", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "9675a230eccdb5464360d4b3dbd866e9f1f5decc3205e7d341806b8e408a0d30", "generation_seconds_batch": 21.133580273017287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06251198053359985, "target_z": -0.47851562914675433, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.6663765907287598, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f87293a6cea3b4ed5b67bf6614f8daa46479a597c55c442a8b79bce4b14f5658.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "f8862f44844c753875716171378f42d9c34620bfb6539bd238337d3e0c7fdf3d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "dd50337edca8fa19623b927ff140fd17b1ffda5ba292e03517b86e4a4ce0cc16", "generation_seconds_batch": 15.820022452040575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.618053913116455, "target_z": 4.312499853335847, "ce": 5.546053409576416, "genuineness": 4.494794845581055, "identity": 0.610536515712738, "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/emo_Confusion/f8862f44844c753875716171378f42d9c34620bfb6539bd238337d3e0c7fdf3d.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "f8c46c56b7be639cf37c2bcc24904a34e8d32caf29f6b5624820258a6adbf4f7", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4e1cb7931c9d9aa133425c6c982895bb8eebc1705ded585eea6231a5f73c9e41", "generation_seconds_batch": 22.45731294795405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07813902199268341, "target_z": -0.4492187521825023, "ce": 6.070269584655762, "genuineness": 4.770940780639648, "identity": 0.7375603318214417, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f8c46c56b7be639cf37c2bcc24904a34e8d32caf29f6b5624820258a6adbf4f7.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "f9c0496a82a2f1c4280d8313630c988565294cdbc718b0532f59be693ff827a8", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c2f45137331de196a4078c2c388ff5c81a2ece5413275ad2cd64f09a40ba7108", "generation_seconds_batch": 15.74047174397856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023965269327163696, "target_z": -0.5507812757535269, "ce": 6.284721851348877, "genuineness": 4.494794845581055, "identity": 0.6573386192321777, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f9c0496a82a2f1c4280d8313630c988565294cdbc718b0532f59be693ff827a8.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "f9d79ce4730f682415e774a3e303d5fcc34c17c95270a4cc55d25d4be8d7ca10", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72aa7edb870113c2a1d7901bbf5466554027056da52664dd0d43daa14369217d", "generation_seconds_batch": 23.289554213988595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08855704963207245, "target_z": -0.4296875008730009, "ce": 6.054384231567383, "genuineness": 3.8797430992126465, "identity": 0.49652114510536194, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/f9d79ce4730f682415e774a3e303d5fcc34c17c95270a4cc55d25d4be8d7ca10.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "fa14f381a4b8dae958ec6e50ca36f97ecb8627f6f4685df00e9c7cae5c72f58e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0e2cf2c1fb81cb39e9369975524fe8aeda9876e4d930113240822990ea92d4c9", "generation_seconds_batch": 15.718381457962096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.401228427886963, "target_z": 2.0312499685719674, "ce": 5.911416053771973, "genuineness": 4.400654315948486, "identity": 0.5576673746109009, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fa14f381a4b8dae958ec6e50ca36f97ecb8627f6f4685df00e9c7cae5c72f58e.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "fa37bb60203e30dce8524316b28e79240a7ccd50d122ead1f24ad904fd13c5ae", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5a7d05396653f110f62b6a49b3d6afad36db5c0a6a90cc8ca97685e4947fec0b", "generation_seconds_batch": 20.009339048992842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.518040895462036, "target_z": 4.124999930159927, "ce": 5.597680568695068, "genuineness": 4.300237655639648, "identity": 0.26004934310913086, "hypothesis": "De deur 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/emo_Confusion/fa37bb60203e30dce8524316b28e79240a7ccd50d122ead1f24ad904fd13c5ae.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "fa65e239b5eda7404cd9bb30ab9947dcbe217b8c3a597972f449ebe33d19531d", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4fffa4c6efe88500550bee62530b434151a81d4a40e12fea50660f8b7adb8a88", "generation_seconds_batch": 22.130511948955245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5595824718475342, "target_z": 2.3281250331740346, "ce": 5.935244083404541, "genuineness": 4.68307638168335, "identity": 0.685076892375946, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fa65e239b5eda7404cd9bb30ab9947dcbe217b8c3a597972f449ebe33d19531d.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "fa8276cd4e087855341c861bcb8fb963c88d34408261c0ffcc08dfc624457a0f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b50c8d205b15c781865a5c9100c9c378a65fd8a249533e18144ba33b58f226aa", "generation_seconds_batch": 20.62079343898222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8803271055221558, "target_z": 1.0546875148410155, "ce": 5.546053409576416, "genuineness": 4.632867813110352, "identity": 0.6980904936790466, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fa8276cd4e087855341c861bcb8fb963c88d34408261c0ffcc08dfc624457a0f.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "fa8c3bdf216fdb94ee62d3735483f54baef1e67869c2f64948453914eac33563", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "73c0af70f395732e67a07ffd3362add0377f34b7408373f4011030d112f5d4ef", "generation_seconds_batch": 21.45652892801445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 5.756534099578857, "genuineness": 4.394378662109375, "identity": 0.6042808294296265, "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/emo_Confusion/fa8c3bdf216fdb94ee62d3735483f54baef1e67869c2f64948453914eac33563.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "fae65e0c5b3c44c95763ccc75ad3ff9560ea7ea0b0e56304f0472b117495e8fb", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7c754cb2d19415002784181583da9e77267322cfaa70ee584a6a3a6edc9d00f1", "generation_seconds_batch": 21.815699794096872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3053826093673706, "target_z": 1.8515625235710247, "ce": 5.202532768249512, "genuineness": 4.482243061065674, "identity": 0.5623480677604675, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fae65e0c5b3c44c95763ccc75ad3ff9560ea7ea0b0e56304f0472b117495e8fb.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "fb12a7345741cb36e48dba6be333f7a0a40554fb00518e7d3f329ee1657858a4", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "85d955a670fe765bb7d507bcc5aa7881390c4a75e4057aebbd25a05745680bc4", "generation_seconds_batch": 15.966613859985955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4845726490020752, "target_z": 2.187499979047978, "ce": 5.478540420532227, "genuineness": 4.419482707977295, "identity": 0.5304170846939087, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fb12a7345741cb36e48dba6be333f7a0a40554fb00518e7d3f329ee1657858a4.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "fb48dc158b48ed80e17461ddbdd101430909ad76741417a306a01f7d1a8e5de5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e8aacadf8b03d40cb83e82e7b0565432e0f9cb6923fd56245aefb1f7981301f5", "generation_seconds_batch": 20.010698296013288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.569869875907898, "target_z": 0.4726562146434631, "ce": 6.086154937744141, "genuineness": 4.319066047668457, "identity": 0.6375935077667236, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fb48dc158b48ed80e17461ddbdd101430909ad76741417a306a01f7d1a8e5de5.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "fb4cc4bac26a2afb95ef6a7636df3a630a8c1cde0e7c2d991f2d99629fe14d2c", "bucket": "emo_Confusion", "phase": "wording", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "67c653bdb86d9086347a2598717bb2d04d08694de3efb298c0554c43c45ef01a", "generation_seconds_batch": 15.829274791991338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2786860466003418, "target_z": -0.07324217844261555, "ce": 6.197352409362793, "genuineness": 4.7458367347717285, "identity": 0.3279186487197876, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fb4cc4bac26a2afb95ef6a7636df3a630a8c1cde0e7c2d991f2d99629fe14d2c.mp3" }, { "id": "emo_Confusion-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-0", "key": "fbb32c493ff779805ed3efdd685e6dbf770230e452e58390a3568ebee29a733b", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La porte est encore ouverte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0653894547483860e589c90e16e5d0387b96e80080d97bd8b1d6cefc9ed1a29d", "generation_seconds_batch": 21.133580273017287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7990664839744568, "target_z": 0.9023437434524931, "ce": 6.038498878479004, "genuineness": 4.7081804275512695, "identity": 0.7854759693145752, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fbb32c493ff779805ed3efdd685e6dbf770230e452e58390a3568ebee29a733b.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "fbfac4899b87f508a0f5d7f4f961f25b9cc28496527bb45ef2118f74bd72b9d1", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "f22c0be3792d8a96feb2ce3951b6b5e0f84f5f260ffa6a3d070d47a96d82cf5d", "generation_seconds_batch": 21.863495272002183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3303859233856201, "target_z": 1.8984376161091212, "ce": 5.696964263916016, "genuineness": 4.444586753845215, "identity": 0.7429835796356201, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereitstellst.", "wer": 0.10526315789473684, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fbfac4899b87f508a0f5d7f4f961f25b9cc28496527bb45ef2118f74bd72b9d1.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "fc18be5aab5f632b97d18422d3b54a9e2a7ede7e76bd7ef6ea3a9ecf1d44d1fe", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d443cf7fe83c0c2662212ed570e075f1bc5936bdb506772a1a8678d209a1a4d6", "generation_seconds_batch": 22.32911538798362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03646692633628845, "target_z": -0.5273437294844786, "ce": 6.38003396987915, "genuineness": 4.482243061065674, "identity": 0.5700370669364929, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fc18be5aab5f632b97d18422d3b54a9e2a7ede7e76bd7ef6ea3a9ecf1d44d1fe.mp3" }, { "id": "emo_Confusion-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-fr-1", "key": "fc1d27c1f6ac59c4c847eb05954535b257ca0a6780893f7e82ee2c76e0fc36f5", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1a7edc8c6c18ecfb8d58ca3da9360cdfc594ae811a519919e5946da057777f71", "generation_seconds_batch": 21.491338068968616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05105215311050415, "target_z": -0.5, "ce": 5.911416053771973, "genuineness": 4.457139015197754, "identity": 0.6152838468551636, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fc1d27c1f6ac59c4c847eb05954535b257ca0a6780893f7e82ee2c76e0fc36f5.mp3" }, { "id": "emo_Confusion-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-1", "key": "fc5113a5a095bf9994b579b9c597b79fa8331fd7ae38a6cfccb60bfda0d41229", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] I found the little box.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] It was beside the old window.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "006e9e77ebbf6202ad0e0c74d86263ddb185845da85b51920c01f5081aa7cf8a", "generation_seconds_batch": 19.816080200020224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18648649752140045, "target_z": -0.2460937609125114, "ce": 6.24500846862793, "genuineness": 4.494794845581055, "identity": 0.3226119875907898, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fc5113a5a095bf9994b579b9c597b79fa8331fd7ae38a6cfccb60bfda0d41229.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "fc71913d65160b47a6a9cb4347da02cbea19a02e45cad2dced7e60023b4b735f", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0d297675d0681457b4f3aa84e739542f9bd23ba65576b0391b2b66dd27064805", "generation_seconds_batch": 20.38351629592944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4563134014606476, "target_z": 0.2597656256547507, "ce": 6.006728172302246, "genuineness": 4.833701133728027, "identity": 0.5775337815284729, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fc71913d65160b47a6a9cb4347da02cbea19a02e45cad2dced7e60023b4b735f.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "fdadf80fb5f424a7e61add0d26582048484db2bb477969b667e82c4364c7ec84", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0f5a391908b604ba77dd5bda32773c0e30a887bdd6fdb5cdc918bf3de2487d5f", "generation_seconds_batch": 24.712741278926842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8012807369232178, "target_z": 2.781250108252113, "ce": 6.133810997009277, "genuineness": 4.425758361816406, "identity": 0.5979485511779785, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fdadf80fb5f424a7e61add0d26582048484db2bb477969b667e82c4364c7ec84.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": "fddfc9008e211f1acac10e8819955e8c543ec43b6c423365c13774db9d456817", "bucket": "emo_Confusion", "phase": "heldout", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e57a0d6356a88e70386f3d15f2fb599b2701ddf3b23227d94ef21b0a1c7f311b", "generation_seconds_batch": 20.89111759501975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4345661401748657, "target_z": 2.0937500174600183, "ce": 5.752562999725342, "genuineness": 4.758388519287109, "identity": 0.7793490886688232, "hypothesis": "Deine Nachricht kam, als ich so ich suche am Al, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.25, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fddfc9008e211f1acac10e8819955e8c543ec43b6c423365c13774db9d456817.mp3" }, { "id": "emo_Confusion-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-1", "key": "fe4ad2dbb42c257d9c68416e2a9ffef6c4f9b57b394a653788a98ceb5b43907e", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] He encontrado la cajita.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4cedcfb109f79d857cd42b4a44b4454e4340cead012a35de941332ccb45dac61", "generation_seconds_batch": 22.11482066300232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5782042741775513, "target_z": 0.4882811709934175, "ce": 5.796247482299805, "genuineness": 4.65797233581543, "identity": 0.6503088474273682, "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/emo_Confusion/fe4ad2dbb42c257d9c68416e2a9ffef6c4f9b57b394a653788a98ceb5b43907e.mp3" }, { "id": "emo_Confusion-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-0", "key": "fe79b9a482aea06b6a91d1f1bb013593fc2a5c6624b656a491aa5bf4ea681db9", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Die Tür ist noch offen.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8623ab62de4961fbb6e287b2f019ee0001f8a643c6d2e352d210da75d02c6d58", "generation_seconds_batch": 20.455518842907622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4448535740375519, "target_z": 0.23828125480150503, "ce": 5.839932441711426, "genuineness": 4.557555198669434, "identity": 0.5161131620407104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fe79b9a482aea06b6a91d1f1bb013593fc2a5c6624b656a491aa5bf4ea681db9.mp3" }, { "id": "emo_Confusion-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-en-0", "key": "fec4fa99be94221b059d5fde3959e86dd35ff34e1d9d986598e4ad33919d0cc0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] The door is still open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] I put your coat on the chair.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a81788ec89734d606f9afd10f03a5718f4cf7fbea85e49a8a0477e4ea82946c1", "generation_seconds_batch": 18.85379116097465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2036762535572052, "target_z": -0.21386717669661373, "ce": 5.792276382446289, "genuineness": 4.400654315948486, "identity": 0.4627872705459595, "hypothesis": "The door is do it for your coat on the chair. Tell me when you are ready.", "wer": 0.2222222222222222, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/fec4fa99be94221b059d5fde3959e86dd35ff34e1d9d986598e4ad33919d0cc0.mp3" }, { "id": "emo_Confusion-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-0", "key": "ff3700d52d66e477765d5e8f33bc38b7da990006b31e949603e13155bae869aa", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] De deur staat nog open.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "07e9c67208b8c0a2d383767aee5e29c8eaf8694e629d05257ea0dfac67f7d7e8", "generation_seconds_batch": 19.690413024043664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9720057249069214, "target_z": 1.226562481666981, "ce": 5.9431867599487305, "genuineness": 4.406930446624756, "identity": 0.5903233289718628, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ff3700d52d66e477765d5e8f33bc38b7da990006b31e949603e13155bae869aa.mp3" }, { "id": "emo_Confusion-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-nl-1", "key": "ffa8134eb656896988ad9673f13780e575db4e27c84382fe7be3a828261723a0", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Het stond naast het oude raam.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ee59db18cb7e4615c140de02d9db31309c2727bd9630422fc1d66825c5d31295", "generation_seconds_batch": 15.182318886974826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6512610912322998, "target_z": 2.5, "ce": 5.998785495758057, "genuineness": 4.425758361816406, "identity": 0.6401571035385132, "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/emo_Confusion/ffa8134eb656896988ad9673f13780e575db4e27c84382fe7be3a828261723a0.mp3" }, { "id": "emo_Confusion-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-es-0", "key": "ffa8c90549a56ba280489a1913d5c59fd6242424a2f2d5e3dfb533372b26bf13", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] La puerta sigue abierta.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "46af5104ed8108e152724e41c8d1672f8d19fcbfb101db1859c6510636214ab6", "generation_seconds_batch": 21.605534713948146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7886484861373901, "target_z": 0.8828125480150502, "ce": 5.927301406860352, "genuineness": 4.507347106933594, "identity": 0.740018904209137, "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/emo_Confusion/ffa8c90549a56ba280489a1913d5c59fd6242424a2f2d5e3dfb533372b26bf13.mp3" }, { "id": "emo_Confusion-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Confusion-de-1", "key": "ffd0aa7fd054e9c105fe2cb13478c69b97490ee9eca2bc998ed38ffa76715103", "bucket": "emo_Confusion", "phase": "grid", "candidate": "manual-de7bf3e68eea", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Employ deliberate, uneven pacing, where pauses are not for breath but for genuine processing of contradictory information.\nSCRIPT:\n(Begin with a halting, staggered delivery, allowing significant, uncomfortable gaps between key syllables.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Mid-phrase, accelerate briefly with a rush of confused energy before immediately decelerating again.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(The final utterance should be delivered with a slow, weighted emphasis on the last syllable, as if confirming a difficult realization.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "37c43ca6c18f8d9813f0403eb37d0b1c20b124191fe93e110400fb9f5e749598", "generation_seconds_batch": 20.557995510054752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.003129243850708008, "target_z": -0.5898437225004712, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.5846971273422241, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Confusion/ffd0aa7fd054e9c105fe2cb13478c69b97490ee9eca2bc998ed38ffa76715103.mp3" }, { "key": "original-emo_Confusion-de-0-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "16a7da396120ac5fcc0cdd836de9407c42962483d08bb8aa6901d230d105ceef", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07918082177639008, "target_z": -0.44726563263875796, "ce": 6.395919322967529, "genuineness": 4.507347106933594, "identity": 0.45410165190696716, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/054927416cbbf314be69c13c.mp3" }, { "key": "original-emo_Confusion-de-0-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "89bdf552cf158ef2a638cf285cf77ae6fc98f7bcdd4719a2281943e716205818", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03646692633628845, "target_z": -0.5273437294844786, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.5078843235969543, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/650e42f79fe422c94b69a801.mp3" }, { "key": "original-emo_Confusion-de-0-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "338ca366bdae58347d5918183990f8f7ef11aad9741c2a78a032c768f375401e", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.268836498260498, "genuineness": 4.425758361816406, "identity": 0.4896857440471649, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/8ba50d80d45a63ec077a1972.mp3" }, { "key": "original-emo_Confusion-de-0-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "a1f0d4233bd9ba3547a88abfa3cbf0d07c227a2d404f344972088f67beff9237", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.5201067328453064, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/fe116df0695cbece92589dd5.mp3" }, { "key": "original-emo_Confusion-de-0-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "3b0f6f5f4c54c84a4cfb3f4d39c7b13daedd76f1a14351e0ffad19edc9a26a96", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10105867683887482, "target_z": -0.40625001047601095, "ce": 6.205295085906982, "genuineness": 4.7834930419921875, "identity": 0.6368380188941956, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/361605136642a596e249ad4c.mp3" }, { "key": "original-emo_Confusion-de-0-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "82ee6f802951642bf8847a3a79081b8492da88be1b0508bf70836da21d9a552c", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05834478139877319, "target_z": -0.48632810732173154, "ce": 6.141753673553467, "genuineness": 4.218649387359619, "identity": 0.6285113096237183, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/de50ad68e6fa43600bed929a.mp3" }, { "key": "original-emo_Confusion-de-0-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e5e081ee73457e671d84322cf383adb149f888340d6f6ac779a72fa17aa1afcb", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.229123115539551, "genuineness": 4.331617832183838, "identity": 0.6893645524978638, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/d76d8320e92f83f02aece1a6.mp3" }, { "key": "original-emo_Confusion-de-0-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "9b128da354502828c02cb206b45014c400f7e6a289b1794a4d1a19f9354b95ef", "task_id": "emo_Confusion-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.028132498264312744, "target_z": -0.5429687417064913, "ce": 6.213237762451172, "genuineness": 4.20609712600708, "identity": 0.43111276626586914, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/ca9b02edd99ad399ca042df1.mp3" }, { "key": "original-emo_Confusion-de-1-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "974e873c04cadf95ca067f365bc13ec6354c24838da1aca7295eba02b73e57e3", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.003129243850708008, "target_z": -0.5898437225004712, "ce": 6.348263263702393, "genuineness": 4.137060642242432, "identity": 0.5835342407226562, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/0750e7f813580d9fe41b0ed3.mp3" }, { "key": "original-emo_Confusion-de-1-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "96a31312ed0ac661a7c6e9feeb6aeffd8ed24ef8f370804a4689386ccbb67bb8", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.538887023925781, "genuineness": 4.2374773025512695, "identity": 0.5964388847351074, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/466f04b856a99ce5a8ac5970.mp3" }, { "key": "original-emo_Confusion-de-1-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "0b80aa08887007798d368d450402a9cbdc58db999dbdfa6dd09bc683024407fa", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.030208468437194824, "target_z": -0.6523437713885223, "ce": 6.403861999511719, "genuineness": 4.042920112609863, "identity": 0.5557600855827332, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/fc4ef64c870c582076702d93.mp3" }, { "key": "original-emo_Confusion-de-1-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "fbbb85fd3e83a0abb98aa1f67f4eb156cbbb9859d318738d5603d6f11bd73de0", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00104561448097229, "target_z": -0.5937500174600182, "ce": 6.3641486167907715, "genuineness": 4.080576419830322, "identity": 0.5605684518814087, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/a29ae231d832d9f4c1310391.mp3" }, { "key": "original-emo_Confusion-de-1-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "57184612323cf40325db6517db095df1f79ae08ad5f049c978c45b23d3aee9cf", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023965269327163696, "target_z": -0.5507812757535269, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.48211753368377686, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/cd09a4d764745747c421ea24.mp3" }, { "key": "original-emo_Confusion-de-1-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "1586a8b5e99419341c4ee55c2189180901364948cceaa502d6df0de3122086b0", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.443574905395508, "genuineness": 4.293961524963379, "identity": 0.6156508326530457, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/a97f9ffa87448ddd9fcbf166.mp3" }, { "key": "original-emo_Confusion-de-1-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "f5b586aa571f5135a76be7f34d6818f8c7a9e78df2785cae8499408e903ba099", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009380042552947998, "target_z": -0.5781250052380055, "ce": 6.427689552307129, "genuineness": 4.507347106933594, "identity": 0.5366182923316956, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/a97e7f216b2d12fb29484b34.mp3" }, { "key": "original-emo_Confusion-de-1-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "6543b804450b92fdb42c127aa9ff510df2cb2cd299bc65f3e941286669a931fc", "task_id": "emo_Confusion-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.019790440797805786, "target_z": -0.632812520079021, "ce": 6.292664527893066, "genuineness": 4.256305694580078, "identity": 0.4797620475292206, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/de/df2b669a150c4ae9f87b3c20.mp3" }, { "key": "original-emo_Confusion-en-0-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "a70c06632a24a3870bc8fde28574b5728c0348392eddf77049e29d01990ffe56", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07084640860557556, "target_z": -0.4628906169247416, "ce": 6.125868320465088, "genuineness": 3.9927117824554443, "identity": 0.15412862598896027, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/8130d3ca723c0af7f305155f.mp3" }, { "key": "original-emo_Confusion-en-0-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "81367065c0c39938b817836c828d6287dfbc095cbb4c8c8d4c7778c97e31c5db", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08230623602867126, "target_z": -0.4414062460714959, "ce": 6.23706579208374, "genuineness": 4.005263805389404, "identity": 0.4022502899169922, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/9bb5e407d2cee6d4e6968393.mp3" }, { "key": "original-emo_Confusion-en-0-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "2d92b59afdd6fa2aa57bf8105d3e64b8c3982d19597241d19f13df2183613662", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08334803581237793, "target_z": -0.4394531265277516, "ce": 6.252951145172119, "genuineness": 4.068024158477783, "identity": 0.5550960302352905, "hypothesis": "The 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/emo_Confusion/en/a53e29e18e42372cdcf9333a.mp3" }, { "key": "original-emo_Confusion-en-0-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "7a67a225a1abdcf287b3c4a300627d02249a49199f3de3445c1fb84282002c61", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06147018074989319, "target_z": -0.48046874869049866, "ce": 6.197352409362793, "genuineness": 3.961331605911255, "identity": 0.4498447775840759, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/1c9e75ae94a49b77de6f469f.mp3" }, { "key": "original-emo_Confusion-en-0-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "38f1c2af0b7aeb21ac8942ec3d4b4fc52b0853237887ca607ee948733b4e4117", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06251198053359985, "target_z": -0.47851562914675433, "ce": 6.086154937744141, "genuineness": 3.923675537109375, "identity": 0.4179539680480957, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/610e3e8199f2a399dd96642c.mp3" }, { "key": "original-emo_Confusion-en-0-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "49e1777f80c583163f7e1d49f0e35d78186b2e6ff893a59783d5bc64b50491fc", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06563737988471985, "target_z": -0.47265627051552145, "ce": 6.229123115539551, "genuineness": 3.9425034523010254, "identity": 0.4049931764602661, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/50bf0cc7be2f9eb4744e18fc.mp3" }, { "key": "original-emo_Confusion-en-0-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e02348364a57ea1e20d5370374eca1e64db9e911c698eb34677961f1647ccb89", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08334803581237793, "target_z": -0.4394531265277516, "ce": 6.268836498260498, "genuineness": 4.068024158477783, "identity": 0.0068999119102954865, "hypothesis": "The 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/emo_Confusion/en/2e96b0feccb179d82cee2da4.mp3" }, { "key": "original-emo_Confusion-en-0-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e68d7766c7bab39d4417e68f8d68d9c92c1752405d5bff3e5168314427242fe2", "task_id": "emo_Confusion-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07501360774040222, "target_z": -0.4550781387497644, "ce": 6.197352409362793, "genuineness": 3.8797430992126465, "identity": 0.3583783209323883, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/0feb20182c64206a956f383b.mp3" }, { "key": "original-emo_Confusion-en-1-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "2af688e2dc3d76310a5824c60f6c6ec7dbe2fb2205139a0eec622b9b7e48e562", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06772100925445557, "target_z": -0.4687499755559745, "ce": 6.403861999511719, "genuineness": 4.632867813110352, "identity": 0.4500035047531128, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/091afb6c9088ffb5b9f0eb23.mp3" }, { "key": "original-emo_Confusion-en-1-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "923332b92755726e336cbf3d0189417bf49852c335554a1e7d90fe0c97f266ce", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07084640860557556, "target_z": -0.4628906169247416, "ce": 6.340320587158203, "genuineness": 4.3504462242126465, "identity": 0.39740222692489624, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/e7b3f2a8329bda3913e9f0d6.mp3" }, { "key": "original-emo_Confusion-en-1-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "a0792a61e9e32c3fa266a24598257707bbd0d1c41c0ed74fd6086ad283bc399e", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13856357336044312, "target_z": -0.33593751134901184, "ce": 6.38797664642334, "genuineness": 4.557555198669434, "identity": 0.5461352467536926, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/7030385e570200f26f88d1cc.mp3" }, { "key": "original-emo_Confusion-en-1-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "c391134a7f6d12bea5166fba352d14bb645777ab5b80f336dcab2e643504279b", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11147670447826385, "target_z": -0.38671875916650955, "ce": 6.41180419921875, "genuineness": 4.482243061065674, "identity": 0.3837040066719055, "hypothesis": "I found the l 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/audio/emo_Confusion/en/6b8f2930fd687ac1ae1714b7.mp3" }, { "key": "original-emo_Confusion-en-1-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "3599ae1a1a7ec0c4d59a650decaeb628ae4235edd9d6090076809e519fe0814b", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04896855354309082, "target_z": -0.5039062390874885, "ce": 6.340320587158203, "genuineness": 4.444586753845215, "identity": 0.551525354385376, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/a555691892654d8c709ae06a.mp3" }, { "key": "original-emo_Confusion-en-1-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "de28a42893e3d9eb17194c72fe354158904a30d5d6fb497eae49907823476df7", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13647997379302979, "target_z": -0.33984375043650045, "ce": 6.372091293334961, "genuineness": 4.582659721374512, "identity": 0.39268097281455994, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/7945495205fa44217052ca59.mp3" }, { "key": "original-emo_Confusion-en-1-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "62c2cbf65e171e02fe2f53690dd0aaa5a9394d32673eef2e986f73753907bd2e", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0927242636680603, "target_z": -0.4218749947619945, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.3326227068901062, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/db5ff2e8fec0402ec6bc25fb.mp3" }, { "key": "original-emo_Confusion-en-1-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "1591b45c65278ec466c7aee225201d62c7afe6a3560e3ce99cae617f8f7af5e0", "task_id": "emo_Confusion-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 6.403861999511719, "genuineness": 4.344170093536377, "identity": 0.4825253188610077, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/en/aa87403a75700a25e33ad461.mp3" }, { "key": "original-emo_Confusion-es-0-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "d382aede8c6e2d88f8cd49d4e4363b3ad8f94eaccf1bcc64a64d751ae3dc1b07", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.021874040365219116, "target_z": -0.6367187591665096, "ce": 5.974957466125488, "genuineness": 4.400654315948486, "identity": 0.6484777927398682, "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/emo_Confusion/es/532292bb9a89026815623366.mp3" }, { "key": "original-emo_Confusion-es-0-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "c0c8ea8139092a8b412b1bc49849d6ef6e36a262a8d8dda21036d98f182f4e01", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009380042552947998, "target_z": -0.5781250052380055, "ce": 5.95112943649292, "genuineness": 4.331617832183838, "identity": 0.5690686702728271, "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/emo_Confusion/es/7d48a998d6585434ed6bb723.mp3" }, { "key": "original-emo_Confusion-es-0-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "4bdb4eb874b3e7a8b51749c619bb46ba0b846dcd2f91133abff6d58c04015b28", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.032292068004608154, "target_z": -0.656250010476011, "ce": 6.125868320465088, "genuineness": 4.055472373962402, "identity": 0.7081985473632812, "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/emo_Confusion/es/0579bc02b4e0038c3a942a0a.mp3" }, { "key": "original-emo_Confusion-es-0-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "75462377a295fdb5ff43723ec61d91fe37a73d6c2beffcb766f1ef4622dcff0e", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0052051842212677, "target_z": -0.6054687347224841, "ce": 6.109982967376709, "genuineness": 4.2374773025512695, "identity": 0.46789586544036865, "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/emo_Confusion/es/ca60c19592c56371dd0f2f83.mp3" }, { "key": "original-emo_Confusion-es-0-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "b564226217f7a494d2a3ba7a4de3c0ff69d01096453eb4ee443cc9cb1dd3ccc7", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00728878378868103, "target_z": -0.6093749738099726, "ce": 6.0146708488464355, "genuineness": 4.030367851257324, "identity": 0.5656920075416565, "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/emo_Confusion/es/333c551e1f0f74901a90a58a.mp3" }, { "key": "original-emo_Confusion-es-0-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "0906649303dfbb19411481e452a7a056d0d86c39a0493e304282b7da62a85f24", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.013547241687774658, "target_z": -0.5703125270630283, "ce": 5.871703147888184, "genuineness": 4.444586753845215, "identity": 0.6881612539291382, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/e516702492368d6b6220a1b3.mp3" }, { "key": "original-emo_Confusion-es-0-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "4cf6b04255cb927df0f52855c0580648aa15a404b85cb3d77436c2b38edda2d5", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06667917966842651, "target_z": -0.4707031509717771, "ce": 6.038498878479004, "genuineness": 3.9550557136535645, "identity": 0.503282904624939, "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/emo_Confusion/es/dbf276d97c67c022ba381fc0.mp3" }, { "key": "original-emo_Confusion-es-0-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "35442c1bf44afc2897e7f192aa435b04323af354bab54474959a04351cdd6347", "task_id": "emo_Confusion-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.009372413158416748, "target_z": -0.6132812687695196, "ce": 6.1020402908325195, "genuineness": 4.055472373962402, "identity": 0.4640897214412689, "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/emo_Confusion/es/ab21fdac0f456335f605cfe7.mp3" }, { "key": "original-emo_Confusion-es-1-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e385a3d924eeef19f8ab95513c7efef0df5a359e8eca6d282d89d848e9f97b4a", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.23706579208374, "genuineness": 4.532451152801514, "identity": 0.5527089238166809, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/c7dc336adae3c3c8dc5cd25e.mp3" }, { "key": "original-emo_Confusion-es-1-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "8e175c8638ef1c48e0610f403fc5c0c0a3b0898637341c9f8a9509d6d7dd55d3", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.1020402908325195, "genuineness": 4.632867813110352, "identity": 0.6530462503433228, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/cf4934a6a9b62daf60af40aa.mp3" }, { "key": "original-emo_Confusion-es-1-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "01770609fa9c2295fe0825a2699bf2d7880fd3295bc7fceabb61bcfb08793d7d", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023965269327163696, "target_z": -0.5507812757535269, "ce": 6.340320587158203, "genuineness": 4.7081804275512695, "identity": 0.43917131423950195, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/5fe3e83c685acb6f52f9e837.mp3" }, { "key": "original-emo_Confusion-es-1-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e426fcb970c924c8829caf31e96cc30d419a9cdc8e936bc32ae5da0c73beeaea", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07084640860557556, "target_z": -0.4628906169247416, "ce": 6.109982967376709, "genuineness": 4.15588903427124, "identity": 0.6629124879837036, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/0cd1f2bfbd6a4168076d14e2.mp3" }, { "key": "original-emo_Confusion-es-1-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "0885d95eb64737f3e0cf06075b00dde754df1cc46342eedf24ba7250573b8b83", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.125868320465088, "genuineness": 4.344170093536377, "identity": 0.6554060578346252, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/d079e0f82b0e0dc0785bfebd.mp3" }, { "key": "original-emo_Confusion-es-1-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "0541804f1181882d692242ff1232df87f1b85d177f36eed587692c5c801c7e2d", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2786860466003418, "target_z": -0.07324217844261555, "ce": 6.038498878479004, "genuineness": 4.7081804275512695, "identity": 0.5102860927581787, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/17313266f73b27e2f4d4b093.mp3" }, { "key": "original-emo_Confusion-es-1-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "fcb320f32057a91b52d14483859816c628941b6e9ee8f61ce1a91057562aead7", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.6617913842201233, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/39a74798f1402f6721713c73.mp3" }, { "key": "original-emo_Confusion-es-1-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "4e334cb70253cda83c25f8e3f5a4794894c85431911bb3bcc3b391f8b297864d", "task_id": "emo_Confusion-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009380042552947998, "target_z": -0.5781250052380055, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.651707112789154, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/es/ef76bb3aa2f7b9b4da650be6.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "bcefe9af5e689d2d5b0248e8a89d9f0f4b9bf7fe96220bfe93690041ca76b80a", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021881669759750366, "target_z": -0.5546875148410155, "ce": 6.038498878479004, "genuineness": 4.331617832183838, "identity": 0.6762048006057739, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/b313125f7c072e1045a76ca9.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "8010c3281de9338610b7a628e58bc6c3e826ff79b5a6bfbcc4b00d54111b692b", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.032299697399139404, "target_z": -0.5351562635315141, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.6082066297531128, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/de61075ebec55be85139bbd4.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "97c013d3d4e6901fe13dac1c092543c92e93dc311249389ca9214aa1b65aa795", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021881669759750366, "target_z": -0.5546875148410155, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.4909361004829407, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/0e5f4e4c760cc3833826689e.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "ccc421d86fbae66d22210d22a7d9652471ea931b01c894724ca6e697aacade64", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030216097831726074, "target_z": -0.5390625026190027, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.7602229714393616, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/3c8007156c989c0644c6cd71.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "12278874aee5a51d4e7513938770cb0eb790da3874646baa2cd7b3447811c2c3", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04480132460594177, "target_z": -0.5117187731345242, "ce": 6.09409761428833, "genuineness": 4.7834930419921875, "identity": 0.6744326949119568, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/37a11fa844d585bfccfd7086.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "a2c0a4d1007380619897642ed90d36b5050e1a2ffedd25ffa2eab76a019872d8", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007296442985534668, "target_z": -0.582031244325494, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.6846057176589966, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/33fc0b3f32210e99f3342ed6.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "0d36fe1ac44c174413288bbf8ea069f543747747e73528399840c2f3f279d0c6", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05834478139877319, "target_z": -0.48632810732173154, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.7707973122596741, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/2ec6eaff5b8d9303b94993be.mp3" }, { "key": "original-emo_Confusion-fr-0-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "bd4d3bf7cfa24123d8dfc1e4668fad820113523d11b3dd7cde0546246014c0fe", "task_id": "emo_Confusion-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.034383296966552734, "target_z": -0.5312500244440256, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.7030553817749023, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/ffa6b3e1fbc7b2773b87070a.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "265a764461a1c1fcb30b9a4fcb231783f87d3a525fe6c3336eb24cd1ec44dd8c", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04896089434623718, "target_z": -0.6874999790479781, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.7845696210861206, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/e020dfffd48e5108c4bc84e8.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "773df667e2e665c9f4fc6de0c22eede4605defcb962ebacc6be34164de1985d6", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00728878378868103, "target_z": -0.6093749738099726, "ce": 6.1020402908325195, "genuineness": 4.344170093536377, "identity": 0.7761585712432861, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/f57a8a8fdecdf1cb5b1d780a.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "6fbdb1ca509cef11b1468c89b23fc20e103213c0a97ba92cb2055c2958b07955", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026048898696899414, "target_z": -0.5468749807939799, "ce": 6.395919322967529, "genuineness": 4.7332844734191895, "identity": 0.6365966796875, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/4ddd37b52c38d2dfc09c6672.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "f2b6836348593dc0096bd28469755fb03dd4d0860a3bd9138618026c93dc852a", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026041239500045776, "target_z": -0.6445312373414868, "ce": 6.165581703186035, "genuineness": 4.344170093536377, "identity": 0.7033774256706238, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/b201db19c38b7c12118174f7.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "178ed0eb7fc9d3c5e19209907803cafe448af63f0ef389b30025f6924bb67f30", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00103798508644104, "target_z": -0.5976562565475069, "ce": 6.23706579208374, "genuineness": 4.7458367347717285, "identity": 0.791359543800354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/5798e157217b7084d2b394ef.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "3f56a483bea5883f3452519b8d0c4f3f1c02bdc1fc2fc0458f61accb4ef65bce", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.019790440797805786, "target_z": -0.632812520079021, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.7412525415420532, "hypothesis": "J'ai trouvé la petite la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/4f66803f270789f750f00681.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "63f0e93faf1b72552e4f3a23f58d2896c5a3b0ae003e7fbce0a41409c7b9755e", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.00104561448097229, "target_z": -0.5937500174600182, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.6912051439285278, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/7545c37c50653af95bc437a5.mp3" }, { "key": "original-emo_Confusion-fr-1-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "dc836b8bf9ac64bceb959f03ff77df748aa7bd7385e1526eee679aa4809dde4d", "task_id": "emo_Confusion-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021881669759750366, "target_z": -0.5546875148410155, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.7420749664306641, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/fr/ed7d2df908294fdb1272a44e.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "cae988d3f6ebae7d8f6cebea37f37fc4d5ee3cc15dd136971573c2c60f9b8cc4", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 1.3928940296173096, "target_z": 2.0156250122220127, "ce": 6.149696350097656, "genuineness": 4.2374773025512695, "identity": 0.6185837984085083, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/0b8ffece0f84e6a543a95791.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "f757de023ebd518ee4325b823d3955196f98d6bff75241d54a28ce662b34f464", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39536795020103455, "target_z": 0.1455078250493881, "ce": 6.0146708488464355, "genuineness": 4.005263805389404, "identity": 0.6460774540901184, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/40c3eb7b9912b6f2f43dd3d4.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "d10c92af320da8e81b0579aee56d79c4cebed1a14393ed6ff8ab69c0f0d3d529", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.37088558077812195, "target_z": 0.09960937609125115, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.6389431953430176, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/21c75488bcf7e112297c4830.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "8d89d7bdb03720d05971056dbbbccb391fcf535d0224631433dd3d9f77b12d4d", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5552846789360046, "target_z": 0.44531254103104284, "ce": 6.109982967376709, "genuineness": 4.231201171875, "identity": 0.6949511766433716, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/0ff480535438de68860ec1b0.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "eed8ee64221d61cc82f7e969f7a2602abef940bd107db4157e29927ecd97dd3f", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9511697292327881, "target_z": 1.1875000907920947, "ce": 5.959072113037109, "genuineness": 4.250029563903809, "identity": 0.5846492052078247, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/43d2c04c939dcb35d16fc328.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "f59b2a7d6ee72a72ce0286e4c00784aca82fe78fc4fa53a5172c242e019585ca", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40995317697525024, "target_z": 0.1728515545338667, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.6639937162399292, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/8b218ffdc650a9724763ba7f.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "d0c163d993c5e9e99846a49cb1dae39df612f9ba116f443de5bb5b7dd0d531b3", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9178320169448853, "target_z": 1.1250000419040438, "ce": 5.871703147888184, "genuineness": 4.080576419830322, "identity": 0.6266778111457825, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/db03fc10338f6e792337e7c7.mp3" }, { "key": "original-emo_Confusion-nl-0-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "8f31bce5332f34b2b227b0c7b4fc7730f1d4ab29e979c6451482cdb7b00709ae", "task_id": "emo_Confusion-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15314881503582, "target_z": -0.3085937539285041, "ce": 6.125868320465088, "genuineness": 4.331617832183838, "identity": 0.7322497367858887, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/55f6b00fa54704ff33885c1a.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1701-scale0.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "db617b22856febeae51c4ca64b1bb0f062003c275f48efacb8479d133a0eef1a", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.23597213625907898, "target_z": -0.15332030322436532, "ce": 6.086154937744141, "genuineness": 4.243753433227539, "identity": 0.5209943652153015, "hypothesis": "ik heb het kleine doosje gevonden het stond naast het oude raam we kunnen er samen in kijken", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/55ee1629ca9ea13e7b6d7c5f.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1701-scale0.5", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "f188c6929ec8c3b5b626674bd4991df043163a7a9dcc461aa93435e9a120da9c", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.1708594560623169, "target_z": -0.27539063787676343, "ce": 6.23706579208374, "genuineness": 4.432034492492676, "identity": 0.7466661334037781, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/5dc6851116abc8fd73c99418.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1701-scale0.75", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "a13d28079d541049884249cb6436ed0244a487f31d95ed26a72aa98674660e07", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.04688495397567749, "target_z": -0.5078124781749772, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.8142057657241821, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/8c9aaddc7bdea530000fef60.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1701-scale1.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "c6852ec72e49f4e22312bd22e436dd58dda5c6e9205613e3eb69f146a56cb26f", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.05105215311050415, "target_z": -0.5, "ce": 6.117925643920898, "genuineness": 4.118232727050781, "identity": 0.7540526986122131, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/5579b0c9a761963f1f9108c2.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1702-scale0.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "58494b927b22d61b626b3d286c9c2e9cccac058aa5b69f50de55b2cfe9e24ea6", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.09376606345176697, "target_z": -0.41992187521825025, "ce": 6.054384231567383, "genuineness": 4.432034492492676, "identity": 0.6183485388755798, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/6212f7f7e2a656c611d0cb26.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1702-scale0.5", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "9eb7e08daf4476004e43041378bcda08673d01a3b4ed686b985ee8e5fc7c0893", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.16877585649490356, "target_z": -0.27929687696425204, "ce": 6.284721851348877, "genuineness": 4.413206577301025, "identity": 0.7555243372917175, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/ecc3e06ac2a475ed1914272f.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1702-scale0.75", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "3c0ddbcf9a12d17effad020e92c53cc22e8be85f8dce9159adf13c161a0adb59", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.05521935224533081, "target_z": -0.4921875218250228, "ce": 6.2767791748046875, "genuineness": 4.482243061065674, "identity": 0.6754841804504395, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/5d52e9c4958d723d7730a345.mp3" }, { "key": "original-emo_Confusion-nl-1-seed1702-scale1.0", "bucket": "emo_Confusion", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": [ "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "70f5d9ce9cd37d2d33d96bc59e121fe04fd91768f05a64d25c71ee0812c9ab69", "mp3_sha256": "e3365dc3b064048bd7663ceca07af8ccc7f4ecb78514422116318459827e3df6", "task_id": "emo_Confusion-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Confusion. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Confusion. 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(Intense Confusion. 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(Intense Confusion. 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": 0.2198241949081421, "target_z": -0.18359373996048953, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.6372548341751099, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Confusion/nl/78443a183bf002b9a16f25a4.mp3" } ], "source_standard_audio_unchanged": true }