{ "complete": true, "report": { "bucket": "vn_S_FORM_reg", "label": "Strong Formal Style", "definition": { "id": "vn_S_FORM_reg", "category": "talking_style", "head": "vn_S_FORM_reg", "pole": "high", "label": "Strong Formal Style", "floor": 4.0, "canonical_dimension_name": "Formal Style" }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/style_s_form/full-v2-sealed/sealed-selected", "mission": "style_s_form", "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": 6.417761464118957, "genuineness": 4.198440403938293, "wer": 0.002884347507331378, "identity": 0.5942383629083633, "voiceclap": -0.10908632488921284, "fitness": 0.7835646669863217, "steering_v2": 0.9007642928336294, "raw_target": 2.6772701048851015, "target_cdf": 0.9007642928336294, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.416093511581421, "genuineness": 4.178482608795166, "wer": 0.0038723118279569892, "identity": 0.5942383629083633, "fitness": 0.7799624983617125, "steering_v2": 0.8968385656828732 }, "none": { "n": 50, "ce": 6.419429416656494, "genuineness": 4.218398199081421, "wer": 0.0018963831867057673, "identity": null, "fitness": 0.7871668356109309, "steering_v2": 0.9046900199843855 } } } } ], "selected_wording": { "id": "manual-b9f763b6bfff", "general": "Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.", "inline": [ "Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.", "Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.", "Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable." ], "cfg": 3, "recipe": "P06", "strategy": "initial-physical-directing", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-3", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.531249989655418, "target_raw": 2.8131591200828554, "ce": 6.195763897895813, "genuineness": 4.330048966407776, "identity": 0.6276315160095691, "wer": 0.014625042999656002, "valid_fraction": 1, "identity_delta": 0.06311904788017272, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.6007812464902311, "target_raw": 2.8832699775695803, "ce": 6.212443518638611, "genuineness": 4.234025573730468, "identity": 0.5906613528728485, "wer": 0.01843137254901961, "valid_fraction": 1, "identity_delta": 0.02614888474345207, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.531249989655418, "target_raw": 2.8131591200828554, "ce": 6.195763897895813, "genuineness": 4.330048966407776, "identity": 0.6276315160095691, "wer": 0.014625042999656002, "valid_fraction": 1, "identity_delta": 0.06311904788017272, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.503320327139431, "target_raw": 2.7849966406822206, "ce": 6.105813050270081, "genuineness": 4.307769000530243, "identity": 0.6731616228818893, "wer": 0.02990690574475404, "valid_fraction": 0.95, "identity_delta": 0.1086491547524929, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.4910156417175837, "target_raw": 2.772589385509491, "ce": 6.202117991447449, "genuineness": 4.280154430866242, "identity": 0.5735707148909569, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.00905824676156044, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.4759766054023074, "target_raw": 2.7574249863624574, "ce": 6.2029122591018675, "genuineness": 4.294275557994842, "identity": 0.5596963539719582, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.004816114157438278, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.4660156239840143, "target_raw": 2.7473809719085693, "ce": 6.172730112075806, "genuineness": 4.296785950660706, "identity": 0.5903080500662327, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.025795581936836242, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.4658203419174056, "target_raw": 2.7471840620040893, "ce": 6.205295085906982, "genuineness": 4.3218900799751285, "identity": 0.5823757819831371, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.01786331385374069, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.4554687608987562, "target_raw": 2.7367461919784546, "ce": 6.152476286888122, "genuineness": 4.27230943441391, "identity": 0.602219246327877, "wer": 0.006458333333333333, "valid_fraction": 1, "identity_delta": 0.0377067781984806, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.4398437660710472, "target_raw": 2.720990949869156, "ce": 6.195366716384887, "genuineness": 4.25975730419159, "identity": 0.5744920432567596, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.009979575127363204, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.6007812464902311, "target_raw": 2.8832699775695803, "ce": 6.212443518638611, "genuineness": 4.234025573730468, "identity": 0.5906613528728485, "wer": 0.01843137254901961, "valid_fraction": 1, "identity_delta": 0.02614888474345207, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.576562522536411, "target_raw": 2.8588493704795837, "ce": 6.10591242313385, "genuineness": 4.265719580650329, "identity": 0.5816855125129223, "wer": 0.02369453044375645, "valid_fraction": 1, "identity_delta": 0.01717304438352585, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.575976617224687, "target_raw": 2.858258581161499, "ce": 6.216414833068848, "genuineness": 4.281723463535309, "identity": 0.6297337263822556, "wer": 0.017648778809769523, "valid_fraction": 1, "identity_delta": 0.06522125825285911, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.5365234566651857, "target_raw": 2.8184765458106993, "ce": 6.124676918983459, "genuineness": 4.274506044387818, "identity": 0.6125382587313652, "wer": 0.01843137254901961, "valid_fraction": 1, "identity_delta": 0.048025790601968765, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.522753966585699, "target_raw": 2.8045922756195067, "ce": 6.1588304281234745, "genuineness": 4.277957892417907, "identity": 0.5997542478144169, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": 0.035241779685020444, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.5203124915026647, "target_raw": 2.8021304488182066, "ce": 6.165184569358826, "genuineness": 4.298354959487915, "identity": 0.619379211962223, "wer": 0.01213235294117647, "valid_fraction": 1, "identity_delta": 0.054866743832826616, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.5199218564351695, "target_raw": 2.801736557483673, "ce": 6.197749519348145, "genuineness": 4.241556787490845, "identity": 0.6058630377054215, "wer": 0.009052287581699346, "valid_fraction": 1, "identity_delta": 0.04135056957602501, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.5177734581429831, "target_raw": 2.7995702505111693, "ce": 6.187424063682556, "genuineness": 4.278585433959961, "identity": 0.6081520169973373, "wer": 0.009298245614035087, "valid_fraction": 1, "identity_delta": 0.0436395488679409, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.514257795736235, "target_raw": 2.7960252821445466, "ce": 6.1715387344360355, "genuineness": 4.273250770568848, "identity": 0.5794666953384876, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": 0.014954227209091186, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.5018554751922732, "target_raw": 2.7835195779800417, "ce": 6.209663534164429, "genuineness": 4.268230009078979, "identity": 0.612214320898056, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.04770185276865959, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.4945312509236233, "target_raw": 2.7761343002319334, "ce": 6.158036136627198, "genuineness": 4.294589328765869, "identity": 0.6176028788089752, "wer": 0.020854618163054697, "valid_fraction": 1, "identity_delta": 0.05309041067957878, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.4847656284174067, "target_raw": 2.766287273168564, "ce": 6.167964506149292, "genuineness": 4.309024214744568, "identity": 0.6055514246225357, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04103895649313927, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.463085943734458, "target_raw": 2.7444268703460692, "ce": 6.136988091468811, "genuineness": 4.2939617991447445, "identity": 0.6145980462431908, "wer": 0.024334150326797387, "valid_fraction": 0.95, "identity_delta": 0.05008557811379433, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.4628906380230902, "target_raw": 2.7442299365997314, "ce": 6.192586779594421, "genuineness": 4.31184846162796, "identity": 0.5805809110403061, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.016068442910909652, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.460156287129661, "target_raw": 2.7414727926254274, "ce": 6.201323723793029, "genuineness": 4.259443509578705, "identity": 0.5704456821084023, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005933213979005814, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.4539062324511554, "target_raw": 2.735170638561249, "ce": 6.124081254005432, "genuineness": 4.30431718826294, "identity": 0.6064062096178532, "wer": 0.030152863777089783, "valid_fraction": 1, "identity_delta": 0.041893741488456725, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.4449218623463593, "target_raw": 2.726111376285553, "ce": 6.138973736763001, "genuineness": 4.311848497390747, "identity": 0.5951004020869732, "wer": 0.046437263501891986, "valid_fraction": 0.9, "identity_delta": 0.03058793395757675, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.423876933047736, "target_raw": 2.7048910200595855, "ce": 6.213634872436524, "genuineness": 4.278585410118103, "identity": 0.5870656043291091, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.022553136199712755, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.3914062531403195, "target_raw": 2.672149670124054, "ce": 6.1485049486160275, "genuineness": 4.274506032466888, "identity": 0.5921031698584557, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.027590701729059218, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.3824218593907645, "target_raw": 2.6630903840065003, "ce": 6.210060691833496, "genuineness": 4.320948731899262, "identity": 0.5976835623383522, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03317109420895577, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.3714844085275297, "target_raw": 2.6520617604255676, "ce": 6.186431217193603, "genuineness": 4.236535978317261, "identity": 0.576483091711998, "wer": 0.02777777777777778, "valid_fraction": 0.95, "identity_delta": 0.011970623582601547, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.3552734610985782, "target_raw": 2.635715681314468, "ce": 6.197352385520935, "genuineness": 4.273878443241119, "identity": 0.5703415885567665, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005829120427370072, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.3240234241536417, "target_raw": 2.604205149412155, "ce": 6.208074998855591, "genuineness": 4.3250281572341915, "identity": 0.5688790164887905, "wer": 0.018421052631578946, "valid_fraction": 0.95, "identity_delta": 0.004366548359394073, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.2893554722366785, "target_raw": 2.5692482113838198, "ce": 6.199735236167908, "genuineness": 4.3168692827224735, "identity": 0.5931540206074715, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.028641552478075028, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.2841796876385436, "target_raw": 2.5640292823314668, "ce": 6.171538710594177, "genuineness": 4.328793835639954, "identity": 0.5645124681293965, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 1.3083008060261978, "target_raw": 2.58835147023201, "ce": 6.177098560333252, "genuineness": 4.307141458988189, "identity": 0.5769020523875952, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.012389584258198739, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.2265624870692722, "target_raw": 2.505931794643402, "ce": 6.155653381347657, "genuineness": 4.326910948753357, "identity": 0.5697698667645454, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005257398635149002, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.412695308666963, "target_raw": 2.6936161875724793, "ce": 6.194175338745117, "genuineness": 4.334128415584564, "identity": 0.5465008050203324, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.018011663109064102, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.2911230669522638, "target_raw": 2.5710305404663085, "ce": 6.240878276824951, "genuineness": 4.323082594871521, "identity": 0.587313680499792, "wer": 0.004679089026915113, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.5998828047507996, "target_raw": 2.882364046573639, "ce": 6.228964214324951, "genuineness": 4.279401307106018, "identity": 0.6227012559771538, "wer": 0.005631469979296066, "valid_fraction": 1, "identity_delta": 0.03538757547736168, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.549218755874245, "target_raw": 2.8312776708602905, "ce": 6.257557849884034, "genuineness": 4.280154490470887, "identity": 0.621495686173439, "wer": 0.016623376623376623, "valid_fraction": 0.98, "identity_delta": 0.034182005673646924, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.5218750152213139, "target_raw": 2.803705997467041, "ce": 6.2515214157104495, "genuineness": 4.290949263572693, "identity": 0.6125825208425522, "wer": 0.006680337206652996, "valid_fraction": 1, "identity_delta": 0.025268840342760085, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.02734374065293109, "ci95_z": [ -0.08320312716127878, 0.13789060846714096 ], "language_deltas": { "en": -0.08515622210657309, "de": 0.17109376163765488, "fr": -0.008984393749555198, "nl": -0.11835939892184628, "es": 0.17812495640497514 }, "languages_improved": 2, "clear_gain": false, "delta_ce": 0.006036434173583984, "delta_genuineness": -0.01079477310180664, "delta_wer": 0.009943039416723627, "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": -0.05066404887655473, "ci95_z": [ -0.17238281007087045, 0.04582034901083332 ], "language_deltas": { "en": -0.2859374922415634, "de": 0.0777344395612762, "fr": -0.04531250923623407, "nl": -0.04101561807282441, "es": 0.041210935606572054 }, "languages_improved": 2, "clear_gain": false, "delta_ce": 0.02859363555908203, "delta_genuineness": 0.000753183364868164, "delta_wer": 0.010991906644080555, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.30875973779853577, "ci95_z": [ 0.19195305002025176, 0.4366894405106133 ], "language_deltas": { "en": 0.5460937557264651, "de": 0.20820302903552795, "fr": 0.126953133959147, "nl": 0.33696290633814324, "es": 0.3255858639333955 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.0119140625, "delta_genuineness": -0.04368128776550293, "delta_wer": 0.0009523809523809524, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 4.0, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793033496.3493376, "example": { "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [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.", "language": "nl", "seed": 1701, "scale": 0.0, "cfg": 4.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "0028c8a51e93dd9e81c52755e46c3e8cc3123fd6ef17d18e05e157e5e9f1b7c4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "215fac70ee51ca584a09fcb6f57e92b429c81980e64755881245cc8726f82166", "generation_seconds_batch": 16.83804164803587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.234375011083481, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.730683445930481, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0028c8a51e93dd9e81c52755e46c3e8cc3123fd6ef17d18e05e157e5e9f1b7c4.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "004bb27c1284e94782d3fa4dbd6674d5a820e88fad8098659c81f34cccacbef1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6615ea21f3917e0c2de295aef8ace4331a0bf0771b5b19d4f9b81ed0718e2fff", "generation_seconds_batch": 22.06421726301778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.230215072631836, "target_z": 0.9531250923623408, "ce": 6.403861999511719, "genuineness": 4.557555198669434, "identity": 0.3854723274707794, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/004bb27c1284e94782d3fa4dbd6674d5a820e88fad8098659c81f34cccacbef1.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "00562f2624941a6b91f98c8ca521ae70f856a44ec258d65766f2834df0b99d4e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "51ad2d0e8b4448fa3bb6d108a9da012e2bbefe91b54f64d3a9a4c5cda9c59d52", "generation_seconds_batch": 24.086284031975083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.624999911332153, "ce": 6.23706579208374, "genuineness": 4.457139015197754, "identity": 0.7482585906982422, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00562f2624941a6b91f98c8ca521ae70f856a44ec258d65766f2834df0b99d4e.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "008cfdddb11a249b506477e8cb0aa0a66c04fd45744571ad17080ae98e820ada", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "89f3b3c0b4ceef6aeb059d5a00270c172c2f97db2d4c3a7af83059a604817ac9", "generation_seconds_batch": 21.605806894949637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.7205065488815308, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/008cfdddb11a249b506477e8cb0aa0a66c04fd45744571ad17080ae98e820ada.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "00d627ca225d01dcee6e38d68362166522723fb1a9be56fbb6e1e3a84b392369", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4dacd1fc96b315be398b29f03b4b8ddee483c67e4535efac7adcb07610946991", "generation_seconds_batch": 15.889512825058773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.165581703186035, "genuineness": 4.306513786315918, "identity": 0.6337913274765015, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00d627ca225d01dcee6e38d68362166522723fb1a9be56fbb6e1e3a84b392369.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "00d7d02825016ae79778a02be656d8f1e4435a3ac4744df479f9f33509f2fde0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9a2e0bf59ecc930a1445814a9fa9d456b96510439f81b3794f70be3a84df7e14", "generation_seconds_batch": 21.868054853053764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.046441555023193, "genuineness": 4.337893962860107, "identity": 0.7316089272499084, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00d7d02825016ae79778a02be656d8f1e4435a3ac4744df479f9f33509f2fde0.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": "00e4fc09e9e5b50750a974e3c5d684de77d46513d878b162b177d89d55b8c597", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1347cc405afd8bdc5cb81531283f972b7d2f3545cb67a824b000b3dabac6969e", "generation_seconds_batch": 17.073276616982184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1829493045806885, "target_z": 0.9062500665008854, "ce": 6.1894097328186035, "genuineness": 4.21237325668335, "identity": 0.6488944888114929, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00e4fc09e9e5b50750a974e3c5d684de77d46513d878b162b177d89d55b8c597.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "00f3da96c6e0ec7cfcf01e443c8506711058418fcc20b0a56b3d7f395854c105", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e5cbf7461465277d5998d312e48cc3d3c650c457c4d82b1e4e28034b8e916e56", "generation_seconds_batch": 15.412823204998858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.6743295192718506, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00f3da96c6e0ec7cfcf01e443c8506711058418fcc20b0a56b3d7f395854c105.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "00f877e6e48eb821232b8b8fe984c1503d805c5bdae254502d3ca76848c02f11", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6f8c40bbf6acbdbbff166f0312e0a0199964fbcd32c2bd5aa76139ac3d1c134d", "generation_seconds_batch": 16.787156021106057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.5528898239135742, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/00f877e6e48eb821232b8b8fe984c1503d805c5bdae254502d3ca76848c02f11.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "01a97fc13d78f107ec51610f3c596312628feeea415b32627b79d34646939f3e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "0923bf71682e71294830e3116df664b0caeb4edcb18a8a322ca8b2c58911a27f", "generation_seconds_batch": 15.65274746506475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.624999911332153, "ce": 6.078212261199951, "genuineness": 4.3127899169921875, "identity": 0.6934252977371216, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/01a97fc13d78f107ec51610f3c596312628feeea415b32627b79d34646939f3e.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": "01c68a1710a53f9fb2c2a88698de6d3099e9a06fdec4ecfa24bf529bc58fa3ab", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4092a69dad697cbc43a1293fd38c2cc89d3c2df611ff8c385739166865d386f0", "generation_seconds_batch": 16.109124653972685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1203866004943848, "target_z": 1.8359376459324985, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.7275363802909851, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/01c68a1710a53f9fb2c2a88698de6d3099e9a06fdec4ecfa24bf529bc58fa3ab.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a08ba3ecd7b643bc14f2411b2959ce1a6565cd0f2df8aad9b2058fef0313abd", "generation_seconds_batch": 13.833044898929074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.197352409362793, "genuineness": 4.607763767242432, "identity": 0.5974724292755127, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "031646e4858651c2122dc495da9d08d25724a292b4d63bb5a132073dc0cb7169", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "081899a9ce283143a025edd8660cb042505401e3bc4ac2c05d3c1237e9b4f49a", "generation_seconds_batch": 15.1014803359285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.7343748928596847, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.5868827104568481, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/031646e4858651c2122dc495da9d08d25724a292b4d63bb5a132073dc0cb7169.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "041ba0599a8668ba12fa44a4be392dc4d22e3b455b18bea3f4714937bb3a0678", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cffcc7f1bfa04640848353c1c5345c5d70bf9b8b318486b9b4cfd4c7e996db75", "generation_seconds_batch": 21.692288791993633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6634843349456787, "target_z": 1.3828125535701576, "ce": 6.181467056274414, "genuineness": 4.193545341491699, "identity": 0.8048743009567261, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/041ba0599a8668ba12fa44a4be392dc4d22e3b455b18bea3f4714937bb3a0678.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "053a42b1c03e2033bf853e3839d0447f9df200653f27ab4678bf4463b7030fe9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8f732d6e48f2ccf724cfc5930b016d855abb54b8876543df8f813ffcef5b6288", "generation_seconds_batch": 21.36141680693254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0781250036944936, "ce": 5.967014789581299, "genuineness": 4.557555198669434, "identity": 0.663506269454956, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/053a42b1c03e2033bf853e3839d0447f9df200653f27ab4678bf4463b7030fe9.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "053d9b8085b94001b7410e271f70466128097505491d4ec83fba97bf5f510412", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "343444ea535ac5438d7154b56f23dbdb52ead59f206e040cb616421ddf8069c3", "generation_seconds_batch": 21.981262775021605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.348263263702393, "genuineness": 4.13078498840332, "identity": 0.5000484585762024, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/053d9b8085b94001b7410e271f70466128097505491d4ec83fba97bf5f510412.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "056f576f4610ef93ecf677310f7fa8221218c783b8dd9759d7d7cea914bea96e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "7992f8bcb91991f5195a368ce6450b0ea3165de72165eea40529e9fe6db42af9", "generation_seconds_batch": 19.58858231897466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.767619252204895, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/056f576f4610ef93ecf677310f7fa8221218c783b8dd9759d7d7cea914bea96e.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89350b2dbd9a17348e802602e5bddb181b4173e98f42f67a897d548d7ba1597d", "generation_seconds_batch": 15.534635560004972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9426817893981934, "target_z": 0.6679688229662493, "ce": 6.38003396987915, "genuineness": 4.65797233581543, "identity": 0.5983490943908691, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "06b0474cbfdd2dca4e47b77ec2412edbc98ff21b8ebf10b43801515ad70dc5d4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ec680f5bd918cfc0caa1562c85a903e55017c630d087eb551939da3001af3297", "generation_seconds_batch": 16.20112233608961, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.605911374092102, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/06b0474cbfdd2dca4e47b77ec2412edbc98ff21b8ebf10b43801515ad70dc5d4.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": "06d50f0cfc4a28cba1637a58adea6ebb43d1622bb0c9b02712cf0753fb52d479", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6d376f0cafa5920293830d73a779eb00dd949ee9461a9616f3237c9d76cdb0b", "generation_seconds_batch": 15.692302871961147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6949949264526367, "target_z": 1.4140626496269921, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.5493960976600647, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/06d50f0cfc4a28cba1637a58adea6ebb43d1622bb0c9b02712cf0753fb52d479.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "077bc78f25692965259c950f33aa4cf589caeee5cd066e7b31309d1f45a05412", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0e22084e7c9c379ab09b1d1d9300b43544fe16bd4ce9eb53b551450ae85bd050", "generation_seconds_batch": 21.590355616994202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.348263263702393, "genuineness": 4.419482707977295, "identity": 0.6418099403381348, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/077bc78f25692965259c950f33aa4cf589caeee5cd066e7b31309d1f45a05412.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": "0799ea36ca2ae2bee6e69c67397e6e62567cd8088ce3c4803046e0a37896d5b0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c77a70f109afd64d40a53101fc050bb18a61aea9358824ab3844a15a835da77e", "generation_seconds_batch": 23.65389902004972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.5981249213218689, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0799ea36ca2ae2bee6e69c67397e6e62567cd8088ce3c4803046e0a37896d5b0.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": "0803c03bb86f26c72fdf4ffaa09e66dfc0620c070ac5a9e99aa27fae3ab479d8", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abed3bd76a01bcea6e1d7aa8c01162b7328ce487bf380cc79d9ed77283ad3db6", "generation_seconds_batch": 16.222576422034763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.24500846862793, "genuineness": 4.293961524963379, "identity": 0.8481904864311218, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0803c03bb86f26c72fdf4ffaa09e66dfc0620c070ac5a9e99aa27fae3ab479d8.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "0907a1f15e0b90f3e6b2e0870c4b53874e6dc3fe5d79e8cef9fd60e0bebabcbc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4facb8060c24a2d5906d2056c48ba0867608e0b39f35c0f6c4829f8e03fd9149", "generation_seconds_batch": 21.930678023025393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.7343748928596847, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.7330130338668823, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfliegen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0907a1f15e0b90f3e6b2e0870c4b53874e6dc3fe5d79e8cef9fd60e0bebabcbc.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "09300475f7158cb9e4a5c31ee669b2744bb27d5165c056ea4669f8168a848755", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "9be24e83e37ac5e42e749985724fabf33bfcafc45c13a1d04849b00ae7976143", "generation_seconds_batch": 22.407797732041217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.073120594024658, "target_z": 1.7890623836234505, "ce": 5.911416053771973, "genuineness": 4.193545341491699, "identity": 0.5107883214950562, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que había.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/09300475f7158cb9e4a5c31ee669b2744bb27d5165c056ea4669f8168a848755.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": "09512c018632b78bc496c574a24f60b7f34a56d9a32f083e268a76eb9aae076f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78ac3870a4eb4804c8e1359bbb3590fee7e72a38d26ce229fe3dd52206c1ac71", "generation_seconds_batch": 15.913670809008181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.3641486167907715, "genuineness": 3.9676074981689453, "identity": 0.47883859276771545, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/09512c018632b78bc496c574a24f60b7f34a56d9a32f083e268a76eb9aae076f.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": "0975102686dbeceebd21723282d5cf362a5a3eecad40a4859a97e76f8c636dfe", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a095bba2408496afb40f861df71f7ac4a2c812be00b2fa03174664ccffea0d15", "generation_seconds_batch": 16.216828600037843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.8176062107086182, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0975102686dbeceebd21723282d5cf362a5a3eecad40a4859a97e76f8c636dfe.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "09cb712ce9e1abaab56fb9e08c1bc2e7bed3176fe17e53b281fc4e1b74782cb3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "814a9cbfb15ac952184b04457c83aabdbc14a69a60e6afabeb79dacdcb9f4cac", "generation_seconds_batch": 15.746389175998047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1514387130737305, "target_z": 0.874999970444051, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.7452062368392944, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/09cb712ce9e1abaab56fb9e08c1bc2e7bed3176fe17e53b281fc4e1b74782cb3.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "09cd47968e830a807645f417522ba433a4ce9c961198a7c5a7c0d5aec15e0af5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e36458ed0945f717264d4583f9934ec212dc6159d4a769370334abbca701690d", "generation_seconds_batch": 15.746389175998047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.7434340119361877, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/09cd47968e830a807645f417522ba433a4ce9c961198a7c5a7c0d5aec15e0af5.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "0a0c97f296130f9a44da97113f1e6434ee00134f108eb02a5aff2d16f06a0f5b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9fc7c71230f20c5181787a108ea3e9319e1b82020962cf7248b326b5c38d144a", "generation_seconds_batch": 25.68102716805879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9781310558319092, "target_z": 0.7031250332504427, "ce": 6.260893821716309, "genuineness": 4.369274139404297, "identity": 0.3425723910331726, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0a0c97f296130f9a44da97113f1e6434ee00134f108eb02a5aff2d16f06a0f5b.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "0a3fcf5f1ca3cbc8af9c8ffa25e42718f8d4898663b5f24ef512d33ed856d1c5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2a6c8ffad37decb01babb4171582529a56389714349e1c7929623a66fa7ddcb3", "generation_seconds_batch": 20.870801943005063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9687500221669618, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.6849536895751953, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0a3fcf5f1ca3cbc8af9c8ffa25e42718f8d4898663b5f24ef512d33ed856d1c5.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": "0a51e8afd7f03046879af3b6557184a21e379b22517a8ada8f7e02673fef4fc0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9f6ac3f91a5f85a5248790b7c3be0f789370b47551f90605664cd79813b4632", "generation_seconds_batch": 23.043408458004706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.252951145172119, "genuineness": 4.168440818786621, "identity": 0.7818286418914795, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0a51e8afd7f03046879af3b6557184a21e379b22517a8ada8f7e02673fef4fc0.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "0a52b01da595dd468767cf83413899e9fd9570e08ce8aa6a270720eb8fbce246", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5ae0cbbdec518142606f790e9b5e0c95d6154a628dbf70918c5410bafcd807ec", "generation_seconds_batch": 22.606556402053684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1278059482574463, "target_z": 0.8515625757371195, "ce": 6.356205940246582, "genuineness": 4.344170093536377, "identity": 0.2350471317768097, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0a52b01da595dd468767cf83413899e9fd9570e08ce8aa6a270720eb8fbce246.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "0a768f80d4ff8b214718daddff6380012b96c6bc32f346333d35911dbf660826", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "6f824b8bb315adf2ad61e7358a576b93d3c0a27cf61b8359451cbef4e2020178", "generation_seconds_batch": 22.432739939074963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.340320587158203, "genuineness": 4.457139015197754, "identity": 0.49974602460861206, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0a768f80d4ff8b214718daddff6380012b96c6bc32f346333d35911dbf660826.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": "0aa1bf73fb9f2688a60c94950d917b4580417dc9dcfc497bf187eb595dd34a32", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2aa9fe2c22aeacacca0ea9646df259a064d597d9dcd52c4cbd17ec5fae66518", "generation_seconds_batch": 15.528371631051414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.205295085906982, "genuineness": 4.607763767242432, "identity": 0.6739890575408936, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0aa1bf73fb9f2688a60c94950d917b4580417dc9dcfc497bf187eb595dd34a32.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "0b4b9bcec24c0b5e3311185c13bd75dc64885b0c9b58a7f91e8b863fa5f6744d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7292507cbfb4d567c68d26659316df81783a36c2c7ed8e903e301cedbde746ce", "generation_seconds_batch": 21.820332500035875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6162185668945312, "target_z": 1.3359375277087022, "ce": 6.221180438995361, "genuineness": 4.3127899169921875, "identity": 0.3630250096321106, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0b4b9bcec24c0b5e3311185c13bd75dc64885b0c9b58a7f91e8b863fa5f6744d.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "0ba56a8e191f473623f3cb3ce0944edf729e80f83d66bb1cb37296bfe151f9c6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "061185ec9b85170346af770d69c968e62ebdae3b31f6501dd625a45e35f0a294", "generation_seconds_batch": 23.061784959048964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 5.744620323181152, "genuineness": 4.469690799713135, "identity": 0.4121238589286804, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0ba56a8e191f473623f3cb3ce0944edf729e80f83d66bb1cb37296bfe151f9c6.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "0bbdb0156255cb7c042cb2ee97dfc1dd0b277a8202291f6eb16231eaa1460254", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f82daecf2fa8a86895df28010ce9c3f10dd360a2446af1ea50392cdd3caf00d8", "generation_seconds_batch": 21.500434682006016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3956451416015625, "target_z": 1.1171875646536387, "ce": 6.38797664642334, "genuineness": 4.344170093536377, "identity": 0.46438127756118774, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0bbdb0156255cb7c042cb2ee97dfc1dd0b277a8202291f6eb16231eaa1460254.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "0c25ccecfba585afcf6e7f348e70adf4dbbd03d5f4e61d68962a1e44f3288415", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c636d77640b74c4a3400eece674a0fb3948df99653f6ad0dbbf19098ca8ba489", "generation_seconds_batch": 21.673605085001327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 5.935244083404541, "genuineness": 4.419482707977295, "identity": 0.6342824697494507, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0c25ccecfba585afcf6e7f348e70adf4dbbd03d5f4e61d68962a1e44f3288415.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "0c6a23cb8cc361e28d1567474f3d8e752da3168e2a07cebc805724d6e956000c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c856877d88d13a5bf0164a7abbae3abf2d2602079d56054c98c6647164ddb7c1", "generation_seconds_batch": 23.66228427691385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072662591934204, "target_z": 0.7968750849733536, "ce": 5.990842819213867, "genuineness": 4.7081804275512695, "identity": 0.7071133255958557, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0c6a23cb8cc361e28d1567474f3d8e752da3168e2a07cebc805724d6e956000c.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "0ca92bd95949246cf7dbd0a5b190fe498836f7d9e5212ba17c0cdafe87c3dccc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b3f28d20eff0cb6804d39f02ac5cba744a513a48231bc571b5063e9052e4f108", "generation_seconds_batch": 15.23185336496681, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.515059471130371, "genuineness": 4.519899368286133, "identity": 0.2991483509540558, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0ca92bd95949246cf7dbd0a5b190fe498836f7d9e5212ba17c0cdafe87c3dccc.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbba0769ca383ddb762d081de586b3ac61346f0cd049b7095db175fab813c596", "generation_seconds_batch": 15.824052672018297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.4843748337477867, "ce": 6.205295085906982, "genuineness": 3.9550557136535645, "identity": 0.7052690386772156, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "0cd908868cdb50bde6e1880445d856aa4e93ee6c64b48aa611b0c58c2a2a91aa", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36e727075615a506a94c2dec02035e29b9afe0c2aa1e8ca04520aebaadaaf8a1", "generation_seconds_batch": 23.674761812086217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.141753673553467, "genuineness": 3.9801597595214844, "identity": 0.6336781978607178, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0cd908868cdb50bde6e1880445d856aa4e93ee6c64b48aa611b0c58c2a2a91aa.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": "0dad55d05b6bfa2e803025bfe0b91a49c489342269f172672f8314a9464f851f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "633043bbb77dbdf7102040612805f82a5ea709ca367f83eadc9c7129ee0767d8", "generation_seconds_batch": 22.375424535945058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.7660411596298218, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0dad55d05b6bfa2e803025bfe0b91a49c489342269f172672f8314a9464f851f.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": "0e01ceaa8f1b65fc29ae3e3b5b0ac0510b965761085300bf1b3ea012e6ace1e3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "640fe96dde60960c2335ec6465f63d4fb8e1ae574b52b5b2facc52c0a9c1b3ec", "generation_seconds_batch": 15.464171582018025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.5644014477729797, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0e01ceaa8f1b65fc29ae3e3b5b0ac0510b965761085300bf1b3ea012e6ace1e3.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": "0e82945f8d82908ee6de0e867abc830c7b398b9bc4d79688a0612f4f49b857d0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "836c9c77d305d9ceb1d3b072b2e665400db40b58409234d404f9e95a0f9d4ca3", "generation_seconds_batch": 14.971647138008848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3641345500946045, "target_z": 1.0859374685968042, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.8258298635482788, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0e82945f8d82908ee6de0e867abc830c7b398b9bc4d79688a0612f4f49b857d0.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": "0e9a9cca7df2f2c34ef9703fcf205d47d940cd96f46e70f75a6ad5e307fc180e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "750aa59d4e7a92cd57ec3655a66344b973ab194538e500b653bb9dbe60d3e10d", "generation_seconds_batch": 14.491685796994716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 1.1093748633037357, "ce": 6.292664527893066, "genuineness": 4.300237655639648, "identity": 0.7395288944244385, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0e9a9cca7df2f2c34ef9703fcf205d47d940cd96f46e70f75a6ad5e307fc180e.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "0ee5f59489393b0935d864763f03f01b28f1d2385a26dba9e1c28216c580600d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a16e2da1820fe4d17b3cdbf4a963517f46111c8b4d33b65e50041aa41caf1f17", "generation_seconds_batch": 21.65600947092753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9687500221669618, "ce": 6.197352409362793, "genuineness": 4.068024158477783, "identity": 0.6636363863945007, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0ee5f59489393b0935d864763f03f01b28f1d2385a26dba9e1c28216c580600d.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "0f859e058e88d2a6d379e481330f2e80efffd028b5dc767d3d065b31bd91ee74", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1bab9bcb18516f5309859a75a8784fce614cc12322156683dd6c01946fdb0f2e", "generation_seconds_batch": 21.916815120959654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8367919921875, "target_z": 1.554687490763766, "ce": 6.284721851348877, "genuineness": 4.149612903594971, "identity": 0.7261260747909546, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0f859e058e88d2a6d379e481330f2e80efffd028b5dc767d3d065b31bd91ee74.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "0f868af7644c57d4ce88dc548eb262032d3f350940d4ea73b63fa07910f286b1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f09bdace6d5d6e635253b8865edd6b5cc1264507f4aad59c964ea29d5a11d7f4", "generation_seconds_batch": 26.478625131072477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.213237762451172, "genuineness": 4.3127899169921875, "identity": 0.6329377293586731, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0f868af7644c57d4ce88dc548eb262032d3f350940d4ea73b63fa07910f286b1.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "0fa70164dc69bd10d93a28e44d01d79468e817907394bc317b078c4e41d25589", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "eb30b260992109c5292abe4bcec7d04847edb88edd329975c8f01f56cdc6abaf", "generation_seconds_batch": 21.543968473910354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 6.292664527893066, "genuineness": 4.582659721374512, "identity": 0.7972660064697266, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/0fa70164dc69bd10d93a28e44d01d79468e817907394bc317b078c4e41d25589.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "101cba5d7dd891c60e0b648de8f3b196f197b930fdb3693820889326e40bf327", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "5ac0f26ccef33344bcca99f907c2ade0f5b4b362982a0b8773fd4f053ec6bc99", "generation_seconds_batch": 22.618444842053577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2459702491760254, "target_z": 0.9687500221669618, "ce": 5.998785495758057, "genuineness": 4.406930446624756, "identity": 0.71857088804245, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/101cba5d7dd891c60e0b648de8f3b196f197b930fdb3693820889326e40bf327.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "107f1e29eae54fac120a5723e285bfed59e500b2b5eb1c6e3688debbd71d1053", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "220d3d9f4e23564cf8292819921a8434c81c6097b3a552c05b1fc19611754e99", "generation_seconds_batch": 22.53474323800765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.281249800497344, "ce": 6.141753673553467, "genuineness": 3.911123275756836, "identity": 0.5216712951660156, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/107f1e29eae54fac120a5723e285bfed59e500b2b5eb1c6e3688debbd71d1053.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "108263b8015b472dfc644dc2569ea8a762f51fb4aeff2dd07070431d466fbdd8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2d7184d435ed833e06266fd584a685157eb44c40ccd94e31f571192e27c15bb9", "generation_seconds_batch": 16.980504830018617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.213237762451172, "genuineness": 4.268857479095459, "identity": 0.6335773468017578, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/108263b8015b472dfc644dc2569ea8a762f51fb4aeff2dd07070431d466fbdd8.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "10b064b3e1c241a797260c52936dca46dbaa96a3a0192bea071b03363b1a308e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "02127e80ab8941d63d5ebe3da9b80bdac773fb7997cf854ad3fb0bd1e2ce2076", "generation_seconds_batch": 21.07948115898762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.7109374981527532, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.855961799621582, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/10b064b3e1c241a797260c52936dca46dbaa96a3a0192bea071b03363b1a308e.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": "10b559de62d9d32d190802e34d4b588ca1985408600895c0c6581cfbbd3fa54b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b59f9e607ed1c2f335fe6fb760347f91a647ee0989c6402a06a4920cc0510cf3", "generation_seconds_batch": 22.071688547963277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 6.499174118041992, "genuineness": 4.15588903427124, "identity": 0.3077167570590973, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/10b559de62d9d32d190802e34d4b588ca1985408600895c0c6581cfbbd3fa54b.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "113c93119d08cc4f3869a6e5a3cd1d18668d86a82330daeabfe31c54bc006d9f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "df8474c1337b48484a7dd453cbcce1399160fb843c82a83fc3aad73cc3bebdbc", "generation_seconds_batch": 18.13981138705276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.5390625609591448, "ce": 6.173524379730225, "genuineness": 3.835810661315918, "identity": 0.44847744703292847, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/113c93119d08cc4f3869a6e5a3cd1d18668d86a82330daeabfe31c54bc006d9f.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "119c28c18ffc3fbde2516f1a60a840d8a3b3e744d45ceca21349de024593ea5f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9c22d94aef3073e5bced7742cc225eb8d426aa6aab0b9f8ec6e038dbe77d8564", "generation_seconds_batch": 22.558813657960854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.046441555023193, "genuineness": 4.394378662109375, "identity": 0.5168749094009399, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/119c28c18ffc3fbde2516f1a60a840d8a3b3e744d45ceca21349de024593ea5f.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": "11af65ef272bce931948115f09fb45bc81db32b437ac21d6bf1cbe996f8b60fd", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ece0a6c11b8940b256ded0787f0f57a18c3d94fbce6ea92913f6bca4afde757", "generation_seconds_batch": 16.222576422034763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.125868320465088, "genuineness": 4.394378662109375, "identity": 0.8477422595024109, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/11af65ef272bce931948115f09fb45bc81db32b437ac21d6bf1cbe996f8b60fd.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "122bda7650d01d58c0d4c6ddb219aa0b51109da288169da1bf97a5443a8207e7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b3f88c89235f54413f6956cd9c571efddc5d26ca0f34f48e67a690b55db4001d", "generation_seconds_batch": 17.107057873974554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.4062501847246818, "ce": 5.9431867599487305, "genuineness": 3.8295347690582275, "identity": 0.7606483697891235, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/122bda7650d01d58c0d4c6ddb219aa0b51109da288169da1bf97a5443a8207e7.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": "1257fa839f1c6171d00e871658a07c346add435482bb01aee94515feec0a78a1", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65001ebfc3ad4af88e2e7f25daf0875a3249fdd5a2bdebf6d4065152e79c7ee5", "generation_seconds_batch": 21.715713114943355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 1.1484374242628805, "ce": 6.260893821716309, "genuineness": 4.281409740447998, "identity": 0.7663999795913696, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1257fa839f1c6171d00e871658a07c346add435482bb01aee94515feec0a78a1.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "716aa3240294ecb7459fa022fbb04d4f89ae82d6dc17ff685fdd4045e4fb1e5f", "generation_seconds_batch": 14.77064359805081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0647850036621094, "target_z": 0.789062620071043, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.6136888265609741, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "12d55a2bf9d9e826d196dec4f1341efd2877630c714fd9af96f1a3c1a5da7d89", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4189757fe988ff14dd66e47cdafbcef680671998f24af81e8dd30032aefe5b56", "generation_seconds_batch": 22.432739939074963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.6230561137199402, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/12d55a2bf9d9e826d196dec4f1341efd2877630c714fd9af96f1a3c1a5da7d89.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "12ef6a4e8df087964fb880a5f167b150d7dee1d686c7dfbf08a46a2101f8bcd7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "65284220000d0d368722f711410cfaa2f32cc01a4b30a08e16ad3ba2a900ac5f", "generation_seconds_batch": 22.248808316071518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.343749992611013, "ce": 6.078212261199951, "genuineness": 4.118232727050781, "identity": 0.5848186612129211, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/12ef6a4e8df087964fb880a5f167b150d7dee1d686c7dfbf08a46a2101f8bcd7.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "130f45e1628d2b734dd20ba88264e4311727e83d3b260acb1bec625478c9e62f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "879e347c8bcd632d2db0a387c44559f18334c3fb6cfbfe3b31669ea2f165896c", "generation_seconds_batch": 16.46867455204483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.046441555023193, "genuineness": 4.174716949462891, "identity": 0.6685136556625366, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/130f45e1628d2b734dd20ba88264e4311727e83d3b260acb1bec625478c9e62f.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "148ae8f81d14d8b7c997298ff00caddfc4f73cf632f5a3e24598db8c09d9cf5f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4c34a602458306de38aa56f955dc3104ccb75f9c3308f2aafa7f5764daed283b", "generation_seconds_batch": 16.654883901937865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.8050767183303833, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/148ae8f81d14d8b7c997298ff00caddfc4f73cf632f5a3e24598db8c09d9cf5f.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "14ca75942a3c1cad91181eeb36f599b2fd5edecc773a268a78dc4237381b3a15", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e5a0ad4234ec6142b5154b01ea10ac1f8abb3775fdbb45ec67cafe66c76ca86f", "generation_seconds_batch": 16.71627200092189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 5.879645347595215, "genuineness": 4.093128204345703, "identity": 0.6479942798614502, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/14ca75942a3c1cad91181eeb36f599b2fd5edecc773a268a78dc4237381b3a15.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": "14df0bae5169b59de5e0ab7608a61e6cce35dcb626a889ef2e671bd579286389", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "332bfb53f17f6a85e30ce347468714f2ab65317d9ab48ddff40ecf45a8c2ab55", "generation_seconds_batch": 15.528371631051414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 1.0468751440852517, "ce": 6.054384231567383, "genuineness": 4.406930446624756, "identity": 0.8054530620574951, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/14df0bae5169b59de5e0ab7608a61e6cce35dcb626a889ef2e671bd579286389.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ea95424ec12de574b5f91e1c80354686ce06dfeb1fa5393ff6a826d8949bb06", "generation_seconds_batch": 14.952156457933597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 6.23706579208374, "genuineness": 4.15588903427124, "identity": 0.6072371006011963, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.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": "1569a2ca64f73ac2660444b253ea528bd14218eed4c51110b9c3fe5cb35a3b1a", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b22132b5aaaeb7691d6e5864ae1f13f6b674957c4aa8da4e6dd1715e78b42201", "generation_seconds_batch": 16.083552113035694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.624999911332153, "ce": 6.332377910614014, "genuineness": 4.400654315948486, "identity": 0.780710756778717, "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/vn_S_FORM_reg/1569a2ca64f73ac2660444b253ea528bd14218eed4c51110b9c3fe5cb35a3b1a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "15dacb1cdb3a076f6c47f0fd53553ef34e365fecbf9854129f3281fc49502d3b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "3c7ce29f61f228c9a809bd0e6c7a52b4b86edca7817a9223a738d00acb519f1d", "generation_seconds_batch": 21.023523634066805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.7343748928596847, "ce": 6.268836498260498, "genuineness": 3.9927117824554443, "identity": 0.7920609712600708, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/15dacb1cdb3a076f6c47f0fd53553ef34e365fecbf9854129f3281fc49502d3b.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "170857bba2948e42ef0a31040d2040f3f4fe191a2034f63e29a41e13a618c409", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bd13c0a30d83d42e96f743a55b6b374355d2bb79eed02e3a79dab527cf31c6ec", "generation_seconds_batch": 22.53474323800765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.054384231567383, "genuineness": 4.030367851257324, "identity": 0.6118814945220947, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/170857bba2948e42ef0a31040d2040f3f4fe191a2034f63e29a41e13a618c409.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "1739ace112cd7f5a2c85eeec4159782df08890dc464e62e48bfb5cd98ad70c69", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7ea31f5789473c11d686be70fc7ed2d84ccd018dfcbe2421bed20dcc6f75b42e", "generation_seconds_batch": 22.04793039895594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.2767791748046875, "genuineness": 4.406930446624756, "identity": 0.5312396287918091, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1739ace112cd7f5a2c85eeec4159782df08890dc464e62e48bfb5cd98ad70c69.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "176d8f252a769f7ae2ce67de6602c388a479daf6748bd11e1e73c36eef91e565", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "3684ff4ac3421008c332f4bdfd23c75f1657748b99c519c4723f4352c61b7514", "generation_seconds_batch": 22.83557626605034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.4692317545413971, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/176d8f252a769f7ae2ce67de6602c388a479daf6748bd11e1e73c36eef91e565.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": "178d0d6dfc28ad7242bdaec58f11054d2d71cef87b341791238291b0770c4d6e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39eda3b1bdc5b5aa868c6f889da6edc271f516f894479596d4f677aff1f84c33", "generation_seconds_batch": 22.56166711600963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.252951145172119, "genuineness": 4.2249250411987305, "identity": 0.6789357662200928, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/178d0d6dfc28ad7242bdaec58f11054d2d71cef87b341791238291b0770c4d6e.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "17f52c66ccd42b94477df0c27d42bdd3c9fde43f6eedede67d6ba54c1e747383", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "998cb52c4146eaaef8168adf12877785edec1ed0af98f9d88155e182ba0a5fe7", "generation_seconds_batch": 24.006448970059864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.022613525390625, "genuineness": 4.394378662109375, "identity": 0.7306169271469116, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/17f52c66ccd42b94477df0c27d42bdd3c9fde43f6eedede67d6ba54c1e747383.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": "1830aaa63dc0046fb8c8bd1f6337a17f9d6d66088255ff37eacde5dc5da06b11", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d37db33f1cd7cea0de7532d8b2d1418e1ec1a1bfc7de5a7cc78a7022421a94bb", "generation_seconds_batch": 16.920426217955537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.7251520156860352, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1830aaa63dc0046fb8c8bd1f6337a17f9d6d66088255ff37eacde5dc5da06b11.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "185f617f9604c3eb114bd1a2d83138d0c32bdb06c80c33b24da52aba0f10a3da", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d712b99db790985cb633b6dec8654b19e8467f128c712b7814636d6a3779687a", "generation_seconds_batch": 20.150674605043605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.521934986114502, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/185f617f9604c3eb114bd1a2d83138d0c32bdb06c80c33b24da52aba0f10a3da.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "1a230652634bfb4949e8d011b75d95753b91d6af44fe96467a08bf6fae38cc7c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bad1cb993ed65fe3d4d6c2a544ef1112b0a071609aabea1f4cbd8f4817303a01", "generation_seconds_batch": 15.65274746506475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.149696350097656, "genuineness": 4.375550270080566, "identity": 0.580641508102417, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1a230652634bfb4949e8d011b75d95753b91d6af44fe96467a08bf6fae38cc7c.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "1a2dd619aa7787067efe77a8bed587e39d329c11f5b825ded229afc87a5e67b6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "972b45aa6cd13c7b1316877fc0b1dd30b221d6193ae8d37b06bad951652acbb4", "generation_seconds_batch": 24.09006033896003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.09409761428833, "genuineness": 4.337893962860107, "identity": 0.7380092144012451, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1a2dd619aa7787067efe77a8bed587e39d329c11f5b825ded229afc87a5e67b6.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": "1acc50dfbb3c1ce73abbf0497767e134651b79ba8be15ed01937b3b47475bef0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1dc0fb35b49e96ada9b06472c94c0232a4616209e6c069573a7262ef8d71bf98", "generation_seconds_batch": 18.71628958999645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.2187498448312675, "ce": 6.157639026641846, "genuineness": 4.425758361816406, "identity": 0.7930448055267334, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1acc50dfbb3c1ce73abbf0497767e134651b79ba8be15ed01937b3b47475bef0.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": "1acee86c69a1a86ebce39b621536deba0b98b69cf654b37464870ba4f47da509", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "267890ce565b208bf0aeeaff145b08aed046f64c6b55b2e2488fd042b018bfd6", "generation_seconds_batch": 15.685634549008682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049029588699341, "target_z": 0.7734374538188296, "ce": 6.221180438995361, "genuineness": 4.519899368286133, "identity": 0.7696959972381592, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1acee86c69a1a86ebce39b621536deba0b98b69cf654b37464870ba4f47da509.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "1ae69f8b97a0a8bc6039a3c2dbb7b3682968536c043afdc59c08e014b4bc1f61", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "21c9c96acf055b94b01d3afd2584aa00bdd9cf7ff18fdc71f2a8c734d1e04a4b", "generation_seconds_batch": 21.78842439490836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.9843749519715828, "ce": 6.197352409362793, "genuineness": 4.425758361816406, "identity": 0.33989980816841125, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1ae69f8b97a0a8bc6039a3c2dbb7b3682968536c043afdc59c08e014b4bc1f61.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "1b2dab3aac5a70031ff6199db3fa1480aa69a67413039d42d75dd1d00cfeedcc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9556da59fe613bdbf2a6628aa1e160e7033710b0689828b1deb1c3b921a5a515", "generation_seconds_batch": 21.2765400740318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.8026711344718933, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1b2dab3aac5a70031ff6199db3fa1480aa69a67413039d42d75dd1d00cfeedcc.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "1b6dd72b152df6edc2460de1ced5529f6ccc2413fe96dde17aa5219c635e8027", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f0654365e71debdaa1493758aef91975ebeb4de9d69953d6f8f87d3a02783c52", "generation_seconds_batch": 22.53672899294179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.070269584655762, "genuineness": 4.293961524963379, "identity": 0.6510419845581055, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1b6dd72b152df6edc2460de1ced5529f6ccc2413fe96dde17aa5219c635e8027.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "1cb5b021f84cf30e71710070d9b3a73fca4ba90e61b9c885923a49c165fc57b5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4aef0bb0e97f721a8c7f02dc0fad9cbb30211a42c1aa72f4845f83ec56392810", "generation_seconds_batch": 26.066171929007396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.2656248706927227, "ce": 6.086154937744141, "genuineness": 3.8420867919921875, "identity": 0.7941944003105164, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1cb5b021f84cf30e71710070d9b3a73fca4ba90e61b9c885923a49c165fc57b5.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": "1cd47ae13cf3cda7f0c270cdb8104b1f572ab8605152f95364dfbabc7aa97fd4", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb67d900987c8feab894a5641a2327c68cf1196844ab1ff004b077129d78af5c", "generation_seconds_batch": 15.983335622004233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9374999261101273, "ce": 6.205295085906982, "genuineness": 4.168440818786621, "identity": 0.6966111660003662, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1cd47ae13cf3cda7f0c270cdb8104b1f572ab8605152f95364dfbabc7aa97fd4.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": "1cf509acc2186284a165b916c61af261a3d93a1c28d856a5d424a0142d5f6acb", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95e2d399e4a0aba838da77f45487533ecebe66dce0e9e7773760be092f3e8cfd", "generation_seconds_batch": 15.987199209979735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.459460258483887, "genuineness": 4.168440818786621, "identity": 0.24458014965057373, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1cf509acc2186284a165b916c61af261a3d93a1c28d856a5d424a0142d5f6acb.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "1cf7f95acbd77e30a66f280c4cd266425a97c99d68888b13dcf7500283d6405c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a1857f52cfb1e2399ca603f7d9759d5c2d8c38337ef678676b1622fb53e1c59b", "generation_seconds_batch": 20.372033560066484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0923566818237305, "target_z": 0.816406365452926, "ce": 6.1894097328186035, "genuineness": 4.18099308013916, "identity": 0.45338550209999084, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1cf7f95acbd77e30a66f280c4cd266425a97c99d68888b13dcf7500283d6405c.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c9711bb2fecef53d63891dcb9079bf8240c1471470234404300041da4605b7e", "generation_seconds_batch": 16.024790919036604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.7265626644049665, "ce": 6.23706579208374, "genuineness": 4.168440818786621, "identity": 0.6686142683029175, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6461d81242dcd64482ec1bb2a7aa901881d31fcfc79dfa176053fae9af807853", "generation_seconds_batch": 16.80658993893303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3326239585876465, "target_z": 1.0546873725399697, "ce": 6.3641486167907715, "genuineness": 4.494794845581055, "identity": 0.5842961072921753, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "1dd86f9b537f63a988526e196b52e2b13948db5dcca22c3e3f5b10f89cc2e707", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9298a1556479045f5edd303ed5a4ff498e65011aab41bec21355f2620f0fef64", "generation_seconds_batch": 21.077634089044295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 1.1328124944582596, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.7362094521522522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1dd86f9b537f63a988526e196b52e2b13948db5dcca22c3e3f5b10f89cc2e707.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "1dfe25e02fba4eef450e30d300ae6129ac0b9ac82d04df657dbcf15c00d5ae90", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cef5018e5b27238257dc2eff99f0628161d83c9f1cfd269c5ed7b6c622ecb66d", "generation_seconds_batch": 24.09006033896003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.157639026641846, "genuineness": 4.168440818786621, "identity": 0.6149750351905823, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1dfe25e02fba4eef450e30d300ae6129ac0b9ac82d04df657dbcf15c00d5ae90.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "1e1c0c51957d32afc155beacdcbb953a4c5a90e8f46767ab2960e449228346fd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e797698959f4401249004cc6e22bc21ab5b79af1bc5c63a256cc0bebe44a48cd", "generation_seconds_batch": 20.3730546340812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.165581703186035, "genuineness": 4.15588903427124, "identity": 0.3532286286354065, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1e1c0c51957d32afc155beacdcbb953a4c5a90e8f46767ab2960e449228346fd.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "606dd221b75f67cda6f698a4f8dab60d1b9a74bb7c4216efb7e672f70402a914", "generation_seconds_batch": 16.024790919036604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.348263263702393, "genuineness": 4.319066047668457, "identity": 0.64055997133255, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "1ef7d479e93a4828427d8bf3423500c733efd75b0556009cc2a0a18871026032", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "993f5cfb2027ceec8bb0364ba4b7f9a1cca99daceed46f1d2be1a7fd9fbdd7b4", "generation_seconds_batch": 23.068096271017566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536556243896484, "target_z": 2.2656248706927227, "ce": 6.427689552307129, "genuineness": 4.281409740447998, "identity": 0.47981610894203186, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1ef7d479e93a4828427d8bf3423500c733efd75b0556009cc2a0a18871026032.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": "1f498d56f80bdec8d9c09e2ba018a83b98ec69c3d2e24d9bb6616ea0c5df6142", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fac6972129d0adfed2b0ffea4872890c4a3c73cbcdbc033fa315fb6bed13a4f1", "generation_seconds_batch": 21.495447322027758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.403861999511719, "genuineness": 3.9927117824554443, "identity": 0.2739107012748718, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1f498d56f80bdec8d9c09e2ba018a83b98ec69c3d2e24d9bb6616ea0c5df6142.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "1f712558268425f690d35bc4df4faa9c91f17f67e46da586e6b2f7131bad92ce", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "79e7f1b0f60dd0a60610413517da5a6593150f21499eb314ecd9b9b20315bb7e", "generation_seconds_batch": 15.58795362408273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 6.41180419921875, "genuineness": 4.15588903427124, "identity": 0.36401206254959106, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1f712558268425f690d35bc4df4faa9c91f17f67e46da586e6b2f7131bad92ce.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "1fdb329acc73808abff28885988bc65268daa479a2b1bdd827a7654a5ccb158e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0a632f2d586d9f82f6727e4f6551a614cd95c0f3801f8b965ca6304766b9c131", "generation_seconds_batch": 22.009892446920276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1203866004943848, "target_z": 1.8359376459324985, "ce": 6.356205940246582, "genuineness": 4.400654315948486, "identity": 0.6484023332595825, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/1fdb329acc73808abff28885988bc65268daa479a2b1bdd827a7654a5ccb158e.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "2025976c6c2cdbf3cdefd8b157210d9b41422711e19cfcc2b3c8ab0aca9826c8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "0b82dbbdf577976e7a871577be506bde682baac3515ca05d554b275644eb9608", "generation_seconds_batch": 22.466701232013293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 5.927301406860352, "genuineness": 3.9550557136535645, "identity": 0.6013477444648743, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2025976c6c2cdbf3cdefd8b157210d9b41422711e19cfcc2b3c8ab0aca9826c8.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "20416ee438c74adc14cbaa0af742dc139f88450e1c5bdd1c1906b1b1df020011", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "48e91881a1500fe5df1b281398331132b3444dc89effd50bd9a072a8f50f4412", "generation_seconds_batch": 23.71613089600578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7187499630550638, "ce": 6.173524379730225, "genuineness": 4.268857479095459, "identity": 0.7086915373802185, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/20416ee438c74adc14cbaa0af742dc139f88450e1c5bdd1c1906b1b1df020011.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": "20a3ec00efec978d5ebd40c322fe661929691058345682f0fd9df098eced7d21", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d24bd94194606311d43e2e43a06c1e9ea7480b9be3d1d6e7c1e82026d842470", "generation_seconds_batch": 15.983335622004233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.316492557525635, "genuineness": 4.231201171875, "identity": 0.6203484535217285, "hypothesis": "Die Schlüss liegen auf dem Küchentisch. Ich suchte in der Schublade dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/20a3ec00efec978d5ebd40c322fe661929691058345682f0fd9df098eced7d21.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": "210fd401709bfb6f65d035fa01b25c19b0c699184fec8d8c8f0c06f20e961db3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d811dab0805afd3dc8156eb6652445a32144cb92043ea6d409bddc879899d6c4", "generation_seconds_batch": 16.566208334057592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2459702491760254, "target_z": 0.9687500221669618, "ce": 6.403861999511719, "genuineness": 4.7081804275512695, "identity": 0.7698140144348145, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/210fd401709bfb6f65d035fa01b25c19b0c699184fec8d8c8f0c06f20e961db3.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": "2114854ce08d49b3da01255c12f1b2fa37153dd198527d04b0d44d45173ae112", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "964986966d8167cda7aa8c815ec0e0b2fb155cb69ecea903ce1f50246a67d0c7", "generation_seconds_batch": 13.53319508396089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1750717163085938, "target_z": 0.8984376015985749, "ce": 6.23706579208374, "genuineness": 4.256305694580078, "identity": 0.6462842226028442, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2114854ce08d49b3da01255c12f1b2fa37153dd198527d04b0d44d45173ae112.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "211e4e2475192b484f18c6e9cc14e2b179772d16757cfa77346fd148d414c9f2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "42c5a04e35d9be758e41c0970767410dffaa597fef71289119035667856d2786", "generation_seconds_batch": 21.832369793904945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.356205940246582, "genuineness": 4.519899368286133, "identity": 0.7600692510604858, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/211e4e2475192b484f18c6e9cc14e2b179772d16757cfa77346fd148d414c9f2.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "217c8978923bfc9d9b0a953c2867c47b20e65e9ce9f9b5341e5af745879003f3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1932ee1e133db8448a487375c478c4fd7a980e3129d0179c96f5b155df50496e", "generation_seconds_batch": 24.660785106010735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.4687501403907581, "ce": 6.197352409362793, "genuineness": 4.293961524963379, "identity": 0.536109209060669, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/217c8978923bfc9d9b0a953c2867c47b20e65e9ce9f9b5341e5af745879003f3.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "22111b0524cae292472d4ec6ab035b7ba3e8d433bd6321bb2eddba82dd4203f1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "483b604d471a5f23766679faa2789cf0e75470fb636e6944428c2ca6a6ff257e", "generation_seconds_batch": 21.987151146982796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.6640624722912978, "ce": 5.935244083404541, "genuineness": 4.124508857727051, "identity": 0.8227812051773071, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/22111b0524cae292472d4ec6ab035b7ba3e8d433bd6321bb2eddba82dd4203f1.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "223a28c52052b9b8375fefeaccdaa3dec91a9859d94fe0e6478f44ff7198309c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6628a052a7c274826c86eb9cdfec917f35db8d41a1ac8ebb830c023a1189cd17", "generation_seconds_batch": 15.79761787399184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 1.1093748633037357, "ce": 6.268836498260498, "genuineness": 4.306513786315918, "identity": 0.7987489700317383, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/223a28c52052b9b8375fefeaccdaa3dec91a9859d94fe0e6478f44ff7198309c.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "2270268586dc292fc7f6ac0b3877b85af15dd54f7e207780feb34e04b826811d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "79d15392a1f10dc9427b1dc1592f4f3a8eab24d808bd4616f2e7c7628bfc71b5", "generation_seconds_batch": 22.86087413295172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.149696350097656, "genuineness": 3.9550557136535645, "identity": 0.20364519953727722, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2270268586dc292fc7f6ac0b3877b85af15dd54f7e207780feb34e04b826811d.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "23c08785cdc6fd3c6eb1c3611258c5aa8244dabaee1b8a30eee7457c8307982a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "eb3262a37e17797eb09ab947b4a556c58dbc2ab7f106d2f3ec4e60aa058311fd", "generation_seconds_batch": 22.81979105202481, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.133810997009277, "genuineness": 4.143336772918701, "identity": 0.7188361883163452, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/23c08785cdc6fd3c6eb1c3611258c5aa8244dabaee1b8a30eee7457c8307982a.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "23d42b526889790a201172873a8453777391b51ac6f3de6fd7ff6ce5fe430c92", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d82d5233b78a76cd6e5cd9ba957456e27d6088a7a3441935c12d8765c3d1faff", "generation_seconds_batch": 22.436698774108663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.986008644104004, "target_z": 0.7109374981527532, "ce": 6.340320587158203, "genuineness": 4.262581825256348, "identity": 0.37505125999450684, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/23d42b526889790a201172873a8453777391b51ac6f3de6fd7ff6ce5fe430c92.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "23f0ed62124b83dec06fde4bf188fbe99d2e23aca1c2f378bcb9e5d9c7dedec9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "48b46f36fbc06f24af6ea6b093473010c332e9344638e5f35f3694ebb18529e5", "generation_seconds_batch": 22.202818003017455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.767404556274414, "target_z": 0.494140592211369, "ce": 6.3641486167907715, "genuineness": 4.607763767242432, "identity": 0.23695649206638336, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/23f0ed62124b83dec06fde4bf188fbe99d2e23aca1c2f378bcb9e5d9c7dedec9.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "240b218851c13f9463394637074385f86b853d953622c0be4c2e525c3cb6e3c0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ae610bff292f7294910c4a23e4ee52ac82de062fa3689f01b7736e85074983c4", "generation_seconds_batch": 17.70522850099951, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9062500665008855, "ce": 6.41180419921875, "genuineness": 4.400654315948486, "identity": 0.4365442395210266, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/240b218851c13f9463394637074385f86b853d953622c0be4c2e525c3cb6e3c0.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": "24484756bf44194ea4aafef2e714298bf000b3ffa600f1029b73a4f0c979181e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eddd3c33a211541c41b2285456c1f7755f885e5db7fca4d01aa0ec86b5fa3b22", "generation_seconds_batch": 15.92403449199628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6886283159255981, "target_z": 0.41601558851687537, "ce": 6.109982967376709, "genuineness": 4.6203155517578125, "identity": 0.7977197766304016, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/24484756bf44194ea4aafef2e714298bf000b3ffa600f1029b73a4f0c979181e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "2487a8d336dbaca83b948f906d54214ef2b38e75c216d992f3b4c4dcb5f9b1ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "649b351ff23f8c6cc7ea283942a1c0771a390d322e77154b0a38afc322cc23e3", "generation_seconds_batch": 23.88473197107669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0962953567504883, "target_z": 0.820312479680285, "ce": 6.006728172302246, "genuineness": 4.68307638168335, "identity": 0.8469802141189575, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2487a8d336dbaca83b948f906d54214ef2b38e75c216d992f3b4c4dcb5f9b1ba.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": "24a39131e18ba4addf9b41205d95101afdc9f2815353f8bee4345b7409de52a7", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af5813f2434b2977138611047311aba5dcdf06533bffbd822cb586826434a34f", "generation_seconds_batch": 22.32076741498895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.356205940246582, "genuineness": 4.344170093536377, "identity": 0.3118349313735962, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/24a39131e18ba4addf9b41205d95101afdc9f2815353f8bee4345b7409de52a7.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "24b948ef2a8c277d740ae5d92da647a424bcd2c74e8ae84d56380640640c3794", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1341345be58f4e55b5df79fb719708f789e5b454dfda1ab99f1bb4f420efd37a", "generation_seconds_batch": 23.385027041076683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.2343750110834808, "ce": 6.1020402908325195, "genuineness": 4.582659721374512, "identity": 0.6699162125587463, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que ha.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/24b948ef2a8c277d740ae5d92da647a424bcd2c74e8ae84d56380640640c3794.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": "24e9b2a87c06b88d9d4b78c286b7e27bebdd720c118d2fff235af97b7b68b086", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ce72e025dafcacaf9c07e1728622451176fe82cda6b3a53db11e113aacb19d5e", "generation_seconds_batch": 22.600771839963272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.23706579208374, "genuineness": 4.193545341491699, "identity": 0.5765249133110046, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/24e9b2a87c06b88d9d4b78c286b7e27bebdd720c118d2fff235af97b7b68b086.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "24fb841754ce455631c476a213c322ca464290c113cd7dad74179581d2e7149b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1bed1f3dbd5148c12dca49c00843886804b099287b388aeb82dae23ad2779109", "generation_seconds_batch": 22.06421726301778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7187499630550638, "ce": 6.38797664642334, "genuineness": 4.281409740447998, "identity": 0.3504921495914459, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/24fb841754ce455631c476a213c322ca464290c113cd7dad74179581d2e7149b.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "25460c2a45366049ed27925e18a35c3a2caf876e0f90c46cc4ad3666002a1351", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e500e17d078738bbc0b51a5f0248720bc006f257889e35e17abe4e120c2fa8c3", "generation_seconds_batch": 21.500434682006016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2105209827423096, "target_z": 0.9335938118827684, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.21505552530288696, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/25460c2a45366049ed27925e18a35c3a2caf876e0f90c46cc4ad3666002a1351.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "254c59fe5ab848f97aa522af4a3d253291c3d230d2d475c3af8c6e0a190ec26b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "4d0464452066d6104cbdbd8294394e27d96d53b28aa5507d5842382604b115cc", "generation_seconds_batch": 27.35881480097305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.268836498260498, "genuineness": 4.218649387359619, "identity": 0.7567906379699707, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/254c59fe5ab848f97aa522af4a3d253291c3d230d2d475c3af8c6e0a190ec26b.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "2647b1a15e479f65f7296117d4d0cef6c83a57e7c1e35f91f4e5a46ed91012de", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c56e4ed69433de05b3826c3bd8a1418e64b041731b07b1416ba477570d6b10b1", "generation_seconds_batch": 21.54012837598566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0844788551330566, "target_z": 0.808593664103023, "ce": 5.8875885009765625, "genuineness": 4.65797233581543, "identity": 0.8197838068008423, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2647b1a15e479f65f7296117d4d0cef6c83a57e7c1e35f91f4e5a46ed91012de.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "266a98973249dfb956f5e97f3941cf966681766e4b91112adc7aacbe4bf02c76", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3e2fe166241ec30956e514aaab67bca78dcad566432b7c367bf59bbac34781ba", "generation_seconds_batch": 23.690930479904637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.340320587158203, "genuineness": 4.306513786315918, "identity": 0.588679313659668, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/266a98973249dfb956f5e97f3941cf966681766e4b91112adc7aacbe4bf02c76.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d2e191ca1920c6fb29308dc5ebba3e983d3192c7d2e721962292cd85808e3040", "generation_seconds_batch": 16.10312503296882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.403861999511719, "genuineness": 4.093128204345703, "identity": 0.10982274264097214, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "2756e04032609d034b8b9edca5f98c1bb597c02d8499b9eae0c358993c95e00a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f5ef86e1a7ef2744fc77d61688f9cd62b5acc74673fc43a02e1f68326fc10a3a", "generation_seconds_batch": 25.411646236898378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868302345275879, "target_z": 1.5859373503730079, "ce": 6.205295085906982, "genuineness": 4.256305694580078, "identity": 0.49604618549346924, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2756e04032609d034b8b9edca5f98c1bb597c02d8499b9eae0c358993c95e00a.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "2772e9b8e06980ef39a767b969487ec8418686cc26247e0a7cf132d6e5569784", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f2623f1f6162046022c5356c087ac5b411504ce07427ff366ba09d4fe9fa671a", "generation_seconds_batch": 21.00680876104161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 1.1015626348490175, "ce": 6.157639026641846, "genuineness": 4.168440818786621, "identity": 0.808076798915863, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2772e9b8e06980ef39a767b969487ec8418686cc26247e0a7cf132d6e5569784.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "27b397a76d66e846ae7ca8c87a3a4626fffd744ccf2bb8de00bf4d18294de913", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "95fb439f291de0e348da080e1c999413c227acf99b7c58bb1bcba66116ecaa01", "generation_seconds_batch": 20.238426917931065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.141753673553467, "genuineness": 4.256305694580078, "identity": 0.6704211235046387, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/27b397a76d66e846ae7ca8c87a3a4626fffd744ccf2bb8de00bf4d18294de913.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "27f5be246e25df0f642e7cb60140ad19a0c88d911f316e3f5025e722eaea6f11", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d3e06209c42bf347b8c07e1da46d4f4ff5a66f93a934a8e14785fcffa84d0ab0", "generation_seconds_batch": 21.525089472997934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.252951145172119, "genuineness": 4.243753433227539, "identity": 0.8084911704063416, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/27f5be246e25df0f642e7cb60140ad19a0c88d911f316e3f5025e722eaea6f11.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "2868d60282bf0a6f14c1590c8e8fdee92170991324d54927fe7057fc094db7fd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "ec514cac2121073d1189355ef5309d536e1358046ca4b53517af664b6e0856bc", "generation_seconds_batch": 22.285901729017496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.181467056274414, "genuineness": 4.15588903427124, "identity": 0.6184965968132019, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2868d60282bf0a6f14c1590c8e8fdee92170991324d54927fe7057fc094db7fd.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "28a9fc8a3ac63c5635613e77428323e70cdb12e899ef315e4b28673403745336", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "3afe37eddb78eeab39529b3d81056f83c48f0d1f6bfb63d4fdb537a2c05f088b", "generation_seconds_batch": 16.83804164803587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2779388427734375, "target_z": 1.9921874168738933, "ce": 6.395919322967529, "genuineness": 4.344170093536377, "identity": 0.6164199113845825, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/28a9fc8a3ac63c5635613e77428323e70cdb12e899ef315e4b28673403745336.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "2916c4bd165d43716bd0f09e27016d6b3dbff569be5947ac08dd04c6e5f3dd27", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3ef3c0d888e7419a1c9ee306b38c322b3b059d79e43e5484e25b93b2429e1ed8", "generation_seconds_batch": 21.295317998039536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879660725593567, "target_z": 0.6054687490763766, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.5880588889122009, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2916c4bd165d43716bd0f09e27016d6b3dbff569be5947ac08dd04c6e5f3dd27.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "2922e0889e04539e52f9186dd74f3c009960cf40f62d59b8f3be75bf79aa287b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "810ce9ba87b823ac205a17730926cec1ced2f40697442be887431e9d7a92a3e8", "generation_seconds_batch": 16.502552989055403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.2767791748046875, "genuineness": 4.017816066741943, "identity": 0.21806249022483826, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2922e0889e04539e52f9186dd74f3c009960cf40f62d59b8f3be75bf79aa287b.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "29d7800f24e66063d0acfb994439d68f5b67709f10743475cdf8c4f3122f7b49", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "03912eae456b00e43662f320ddca4cc5b39a7e07d421b41109535589117f6ba9", "generation_seconds_batch": 21.392400279990397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 1.179687520319715, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.21067985892295837, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/29d7800f24e66063d0acfb994439d68f5b67709f10743475cdf8c4f3122f7b49.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "29e9bd54ccc3c71a45ab74aacf68458f67b35366ad1263e5cb1bb06d403af8d8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "991e569a5d740b75ddf2d391a844e7dee38f65c80c1c2605e52e2ac312806fc5", "generation_seconds_batch": 21.969054025015794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.117925643920898, "genuineness": 4.325342178344727, "identity": 0.6226814985275269, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/29e9bd54ccc3c71a45ab74aacf68458f67b35366ad1263e5cb1bb06d403af8d8.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "2a8b6724f187ff39e689552cd37693e0b0a501bc274718a6a29e709aac441fd4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1a35a20c8b9e55120a30a32faef1a04fe3f540366b5dcb4b6dd49d9f8db0dc9a", "generation_seconds_batch": 22.334150686045177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4609374390408552, "ce": 6.260893821716309, "genuineness": 4.325342178344727, "identity": 0.7300925850868225, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2a8b6724f187ff39e689552cd37693e0b0a501bc274718a6a29e709aac441fd4.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "2ab6fabe26edc1bac96fc42911191cc012a51fc4aab326f3c13f45217aee6282", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ea19bf6c436855c5758f4e1ee7a77b941abebdcb6484f595ad341667f21b69f", "generation_seconds_batch": 25.1623691880377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.234375011083481, "ce": 6.403861999511719, "genuineness": 4.218649387359619, "identity": 0.7159762382507324, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ab6fabe26edc1bac96fc42911191cc012a51fc4aab326f3c13f45217aee6282.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "2ab9d4d768c17648146f9f8ab8c74991c08cce111ad631e00530461e40ac559b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "dd67a37df46a288d373752df3d425b2b61c3d94cadf7bc7bbe6315c0ff6187e0", "generation_seconds_batch": 21.00680876104161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.149696350097656, "genuineness": 4.425758361816406, "identity": 0.6738085150718689, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ab9d4d768c17648146f9f8ab8c74991c08cce111ad631e00530461e40ac559b.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "2b0ef99cffbe319c7fe745f0de4639f9c14536110558eb2b9888dc83d12010d9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a9856ce6195e8ce1b9a023a4e97eac80757d6f16a794c907a3a4a81765cb72e", "generation_seconds_batch": 16.552627812023275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.8081186413764954, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2b0ef99cffbe319c7fe745f0de4639f9c14536110558eb2b9888dc83d12010d9.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": "2b6e937febb64016040bf4746c01c8368f7f5640a1c6c24fee312de1457ddd48", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f52d5fcea1a3c90d471fe5485dc1af1aa81be8d936b104698359f1405b532d5", "generation_seconds_batch": 22.375424535945058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.7299060821533203, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2b6e937febb64016040bf4746c01c8368f7f5640a1c6c24fee312de1457ddd48.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "2b9208f43f763324b120f79d9bf6206570315b3a9e30f34216a3032f0b5a8a56", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ac270d5385373e598a7cf540d43c8301fe05167e26bfa8f896fc9f03fa233b94", "generation_seconds_batch": 22.79169623204507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.0305562019348145, "genuineness": 3.9927117824554443, "identity": 0.6953853368759155, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2b9208f43f763324b120f79d9bf6206570315b3a9e30f34216a3032f0b5a8a56.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "2ba1a5671f721170c07d9a35c1bc2d697643a9df82f3f8e64134dd18ab33991c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e243c2a51a1f6c73b29a11b676e77433f70b6a9eb366ac72ef71fb485e7e176c", "generation_seconds_batch": 26.478625131072477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.316868782043457, "target_z": 1.0390624427353488, "ce": 6.078212261199951, "genuineness": 4.319066047668457, "identity": 0.6325379014015198, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ba1a5671f721170c07d9a35c1bc2d697643a9df82f3f8e64134dd18ab33991c.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "2ba58e4a88486c8dfeba506694d5fce3d0c6b3be3d77bcb075566e5a437146c2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a83f0757d22569ec5eef6925eabd0b60d91caff285b0da40f460e0f2cd4988fa", "generation_seconds_batch": 22.006949549075216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.805281639099121, "target_z": 1.523437631154524, "ce": 6.221180438995361, "genuineness": 4.306513786315918, "identity": 0.7966002225875854, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ba58e4a88486c8dfeba506694d5fce3d0c6b3be3d77bcb075566e5a437146c2.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "2bccf0641319aa01a7e78cffbb295823b182fe4c8afa5c31a24e2ed0c320f514", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "37cf4fbaae914ad68b53cc3dd7c65cc069f95197f1a99a1d1292fa182e22eac4", "generation_seconds_batch": 20.473212865996175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.0305562019348145, "genuineness": 4.413206577301025, "identity": 0.6962078809738159, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2bccf0641319aa01a7e78cffbb295823b182fe4c8afa5c31a24e2ed0c320f514.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": "2c94740770518726e4482b9962d37e07d5b7d4a04e380b9224e5cd79dd69074a", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e5eee6a71659562f12e67e8f903352d96334a9be08057f02913c0dc72407e0e", "generation_seconds_batch": 15.614241737057455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.590793251991272, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2c94740770518726e4482b9962d37e07d5b7d4a04e380b9224e5cd79dd69074a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "2ce5c08bebf14a02bbe6adc6f55bc69dd130e03abb6e54a0a461d976a6f6dca9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "331bc719f5fa4023c611f998562ffc4c5b202ad8dd16d24e966b8b15493ae176", "generation_seconds_batch": 21.360278938082047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.3671873873179443, "ce": 6.181467056274414, "genuineness": 4.394378662109375, "identity": 0.7138444781303406, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ce5c08bebf14a02bbe6adc6f55bc69dd130e03abb6e54a0a461d976a6f6dca9.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "2d90d5a37ce8d8952720c0a8252f6805f25f0d6cee0a722761186dd760edcbf0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "300c24c178bd0694a4fbad0b343eb133c8d2dff6d563a23865f7f0f317e00c6f", "generation_seconds_batch": 25.17194919893518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.017977237701416, "target_z": 1.7343748928596847, "ce": 6.229123115539551, "genuineness": 4.281409740447998, "identity": 0.6307568550109863, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2d90d5a37ce8d8952720c0a8252f6805f25f0d6cee0a722761186dd760edcbf0.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "2e0843dbc8245d80b36cf392ca6b945d0a7d14cd2416c209672a03fd4e1aa5ed", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1d28541536c81323fd7b2d20194ee493ceece5ece8fc72aac72336abdf2bed37", "generation_seconds_batch": 15.580986023065634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0781250036944936, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.6316215395927429, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e0843dbc8245d80b36cf392ca6b945d0a7d14cd2416c209672a03fd4e1aa5ed.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "2e09e4cc2db189fddb309962df910b16580480c3b3a3c7bbb3ceeb84497e2839", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e561fd0a4abc257696ba7ee9a9c853a59860237a5fde42e8c73a6807cdac82c9", "generation_seconds_batch": 22.698530990979634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.229123115539551, "genuineness": 4.306513786315918, "identity": 0.7105373740196228, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e09e4cc2db189fddb309962df910b16580480c3b3a3c7bbb3ceeb84497e2839.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "2e3c24f126f9ddc4d8b3edc82cd915875b9d95a9d9d302327c4a02d334e02160", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e31dbe0343881d1cfa82a14d7b5e2d53761658949718e67cf9d3a7196c5f4178", "generation_seconds_batch": 16.71627200092189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.316492557525635, "genuineness": 4.494794845581055, "identity": 0.753880500793457, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e3c24f126f9ddc4d8b3edc82cd915875b9d95a9d9d302327c4a02d334e02160.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "2e64856163f238ca0b070e5e7ced4a78ef55d8f696cbd74bd13631497b646266", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "94d4e7bcea9ff86c91ccba3d8fe21d567ae33e6817131cef5e0a0e1663fcc78e", "generation_seconds_batch": 21.07948115898762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.0146708488464355, "genuineness": 4.218649387359619, "identity": 0.5815199613571167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e64856163f238ca0b070e5e7ced4a78ef55d8f696cbd74bd13631497b646266.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "2e6feb7655eae12a6a2e66ba4043a13e5b57b78037e4457b827fd56c41bd57ca", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7cf0ce0fae4a46184f1adfec857f6b294489046076b00f44095df965f34888d1", "generation_seconds_batch": 21.618538154056296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.3671873873179443, "ce": 6.395919322967529, "genuineness": 3.9801597595214844, "identity": 0.25009825825691223, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e6feb7655eae12a6a2e66ba4043a13e5b57b78037e4457b827fd56c41bd57ca.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "2e7361fe49b39bd0d076a0fe41736c54f1aaa2feb5447e9062bc77a728f7dd4e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "22c2c4b432471a6457e3242c4c0377da1c6903d02bfe14344790133c4dabb784", "generation_seconds_batch": 20.725853197975084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.300607204437256, "genuineness": 4.256305694580078, "identity": 0.5825259685516357, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e7361fe49b39bd0d076a0fe41736c54f1aaa2feb5447e9062bc77a728f7dd4e.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "2e76ee4f5152c0751407abc9e7448119d4e3336bd5e58191f2be1cdea4ec066e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "09cf2fb60b4de14952436d7c9eb6ca60cc6700c23abe2e5d0896dbcd06c469e2", "generation_seconds_batch": 21.428563820081763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.221180438995361, "genuineness": 4.017816066741943, "identity": 0.3059196174144745, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2e76ee4f5152c0751407abc9e7448119d4e3336bd5e58191f2be1cdea4ec066e.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": "2ec7cb01ac73e352b494104ee11f93a4a5411bc752eb6023c3906848aabdbe18", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d986361b1e09cb5d103400e35e35aaa95c86ff7e89954a80c0ce4ee57d096662", "generation_seconds_batch": 22.56166711600963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.2767791748046875, "genuineness": 4.406930446624756, "identity": 0.627589225769043, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2ec7cb01ac73e352b494104ee11f93a4a5411bc752eb6023c3906848aabdbe18.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "2eef04e846911bdf60002442d2f3c34c9445660ec274ee67eff33137d86ef86b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6ca296471978e2e986a8c9ed7c763f11d7bce9410c30df3930fe83ffde6a9082", "generation_seconds_batch": 21.633461110061035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.1874999852220254, "ce": 5.982900142669678, "genuineness": 4.168440818786621, "identity": 0.5554876923561096, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2eef04e846911bdf60002442d2f3c34c9445660ec274ee67eff33137d86ef86b.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "2f50eac81c733333323da9decfa23fbfb27afdc10ed5a5bacbccc96240fe49b2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8c5478ab9515e612d1b9b73553be787c23b32de5fa6ed9f24c27fbea3263c6a5", "generation_seconds_batch": 16.09147557907272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.078212261199951, "genuineness": 4.356721878051758, "identity": 0.5947786569595337, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2f50eac81c733333323da9decfa23fbfb27afdc10ed5a5bacbccc96240fe49b2.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "2f8c0c798f07af2de42678e08f1235b36fde7e16e2af1c7ad9a3cd112f02259b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3185a2bf678c6a31c48434ee0bb175f872960087dfe286c8a3e2d64b2bc09940", "generation_seconds_batch": 24.51028929289896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.190826892852783, "target_z": 0.9140625314031958, "ce": 6.054384231567383, "genuineness": 4.582659721374512, "identity": 0.590225875377655, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2f8c0c798f07af2de42678e08f1235b36fde7e16e2af1c7ad9a3cd112f02259b.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "2fdad68d61e171a0e28bd33ec8ac83f2b2e46f78d55dc93d4d2f6bbc93f0e2ad", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a5320a3012baceaabb9cb531bb47dad8a80000c1d797eb8b352be50962c25eb3", "generation_seconds_batch": 23.083264735061675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.835937409484906, "ce": 6.086154937744141, "genuineness": 4.7081804275512695, "identity": 0.7268080711364746, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2fdad68d61e171a0e28bd33ec8ac83f2b2e46f78d55dc93d4d2f6bbc93f0e2ad.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "2fe500f63f839cfd16169c5f634c0d7693722aef39e62011fef753af8cbc26b4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cc7e88c7627e732e610eb6f313e2daa5adb02629aa4c326c4311393de6dfe324", "generation_seconds_batch": 20.66833991999738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.5465068817138672, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinfahren.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/2fe500f63f839cfd16169c5f634c0d7693722aef39e62011fef753af8cbc26b4.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "3008f74fd74436b5c3bab9367202bc54a5e22c39fec3f89de06370444b9b74ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8ad9009d5c15ec4f14981fe49af418fa845ecf0b691489c16d87685b11d96312", "generation_seconds_batch": 22.407797732041217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.125868320465088, "genuineness": 4.231201171875, "identity": 0.7196095585823059, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3008f74fd74436b5c3bab9367202bc54a5e22c39fec3f89de06370444b9b74ba.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "30c5fbb3e2242fac0e22d9f5eb9663a8d21e5be0cb3f9b7b0a8a3778b8db90f4", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b183e3e0fc943fd280133de60d2421b282acb5c93b44e10c5d1898a9ec59741a", "generation_seconds_batch": 19.721828431007452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.7193129658699036, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/30c5fbb3e2242fac0e22d9f5eb9663a8d21e5be0cb3f9b7b0a8a3778b8db90f4.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "31b64c543db0515b1c0f5f9429ce362851c959cf0e4adc9921fdad1334d2ce21", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2dd06ada71f720beb287450cbde968d38591929c84f3b6108c642f26740d184b", "generation_seconds_batch": 16.67316015798133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072662591934204, "target_z": 0.7968750849733536, "ce": 5.982900142669678, "genuineness": 4.758388519287109, "identity": 0.8458621501922607, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/31b64c543db0515b1c0f5f9429ce362851c959cf0e4adc9921fdad1334d2ce21.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "31bcc6cd538b8f6abc340bd0aa80d84a369374b8eaeadf42fbeceb5ed1fa00bf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "08a739570f13cbad128721858de91d22fff73badb46458b81602e093cd75f985", "generation_seconds_batch": 20.37553729210049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.229123115539551, "genuineness": 4.262581825256348, "identity": 0.632773220539093, "hypothesis": "Ik heb heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/31bcc6cd538b8f6abc340bd0aa80d84a369374b8eaeadf42fbeceb5ed1fa00bf.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "31c01faa856b864d6a9225289d1f53f5c291eeb83d8112e82dc7a8d6fbfefd3c", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "afb3e8c6e623e24f4c15c814553e7c889de09c68069ae07b2b541ca0ac1a5645", "generation_seconds_batch": 15.166684727999382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.356205940246582, "genuineness": 4.243753433227539, "identity": 0.47880280017852783, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/31c01faa856b864d6a9225289d1f53f5c291eeb83d8112e82dc7a8d6fbfefd3c.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "320c7b3eae3ad30d73a1da8326b5fd8cb915c60c828fff96a5d22fd9881bfe57", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "365abf9a60eb315ec46e4b069d0f59b76ac78a5493d106dd280d8e329c65f4f9", "generation_seconds_batch": 21.451275691972114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.332377910614014, "genuineness": 4.557555198669434, "identity": 0.544113039970398, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/320c7b3eae3ad30d73a1da8326b5fd8cb915c60c828fff96a5d22fd9881bfe57.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "32177a922d20a6121cb98df001f7e608134440da5cea4873b07d41c16df77ea0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d5a98f67322b28e5258697f345401e09200ca64f9c3a7f003a1707d9656f05f4", "generation_seconds_batch": 25.377015238045715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.260893821716309, "genuineness": 4.218649387359619, "identity": 0.7973684072494507, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/32177a922d20a6121cb98df001f7e608134440da5cea4873b07d41c16df77ea0.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "327b1036e09295015be916715f25a22db797b66757c5b1fe6104265ba6db2a7b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3adb9d8445120d3234676fac479000be7df3000d9b193d9989d124f460ff13ce", "generation_seconds_batch": 22.19881454401184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.445312509236234, "ce": 6.403861999511719, "genuineness": 4.362998008728027, "identity": 0.311463326215744, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/327b1036e09295015be916715f25a22db797b66757c5b1fe6104265ba6db2a7b.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": "3292e377ba60eb2e69dba1c152f055cfe81ccd37a6acbbb792a3263a25445a78", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "446019a43625a2d75f32adab578b10730475b567d63cfb2546cefee540656566", "generation_seconds_batch": 15.69572977302596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.443574905395508, "genuineness": 3.9299514293670654, "identity": 0.41906219720840454, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3292e377ba60eb2e69dba1c152f055cfe81ccd37a6acbbb792a3263a25445a78.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": "329ed4df5eb87f9c2dc81d73205e4abb6afcd5211fd6686c08548c788996586d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3c597244c7eb0459e4857c9a9ca54e4b543b7ff8b3a93d2fcd101eb7e9857eed", "generation_seconds_batch": 16.614388455986045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.41180419921875, "genuineness": 4.262581825256348, "identity": 0.2816556990146637, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/329ed4df5eb87f9c2dc81d73205e4abb6afcd5211fd6686c08548c788996586d.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "32bcf2f4827f27bfb1dc8eed9c98c0b6c7a1b3f16371731809d5eaa72b18de9f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7c9c13d19f5aa5d017eb7dce251e6a0f42bc92ccbfb96982f31354561de1a035", "generation_seconds_batch": 22.436698774108663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5822803974151611, "target_z": 0.31054683944049877, "ce": 6.403861999511719, "genuineness": 4.607763767242432, "identity": 0.49968868494033813, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/32bcf2f4827f27bfb1dc8eed9c98c0b6c7a1b3f16371731809d5eaa72b18de9f.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "3318abd3fc1f6154b1d845951672e9aa13569e51966346379a6344d980623f1b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "700fff3811ca222a81bc6ec367d97d9d6213d2cb5931b5fb2b083ccb354e9765", "generation_seconds_batch": 21.994545589084737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3011136054992676, "target_z": 1.0234375129307278, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.5869234800338745, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3318abd3fc1f6154b1d845951672e9aa13569e51966346379a6344d980623f1b.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "33e050cfc1c493d9eae62d346541959e189e9f0574558dd61d19b6a31e1a1cbd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b04b10c8a1f778569beeccc75b08d41e8bf3646233d971a044c31e8adb7dd00b", "generation_seconds_batch": 24.660785106010735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.7499998226643056, "ce": 6.173524379730225, "genuineness": 4.2751336097717285, "identity": 0.6493536233901978, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/33e050cfc1c493d9eae62d346541959e189e9f0574558dd61d19b6a31e1a1cbd.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f70c7eea7ce33a45cc9c1cd43f98f0dd504871c741d2fa7fbdc8537a7f1a7b2d", "generation_seconds_batch": 18.72672589798458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0962953567504883, "target_z": 0.820312479680285, "ce": 6.0305562019348145, "genuineness": 4.2751336097717285, "identity": 0.5999791622161865, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "340aca13087b508077ffcca144a92df085d08c8d3aa5ab9ad0d5330c426644c4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "88d1f239efe9252c9f9bc191e12c71a755246d6d97f24497982a5745278fbec3", "generation_seconds_batch": 16.313538299989887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 6.062326908111572, "genuineness": 4.2751336097717285, "identity": 0.5659623146057129, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/340aca13087b508077ffcca144a92df085d08c8d3aa5ab9ad0d5330c426644c4.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "342a04b0da90431dc22e03ee5a0bbd5ef21f9588a902320608bf20cb499d5989", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f46580ae6bb7fbd549efef9c7a3742ad69c3edfdfcda9b4d2c1d8e9f2b1edfa9", "generation_seconds_batch": 16.980504830018617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.4375000443339236, "ce": 6.316492557525635, "genuineness": 4.017816066741943, "identity": 0.7022014856338501, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/342a04b0da90431dc22e03ee5a0bbd5ef21f9588a902320608bf20cb499d5989.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "34434cd27d70beabaac4b9dd8dd35f850428b8fd6f0325b14137b000d66540e4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "aad0598f36f78787af0fa6d34a94f006667523d95d72d52474f038c70e3dbdf8", "generation_seconds_batch": 21.208012967952527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3011136054992676, "target_z": 1.0234375129307278, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.7641205191612244, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/34434cd27d70beabaac4b9dd8dd35f850428b8fd6f0325b14137b000d66540e4.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "346c55e1fe5cc2c9dfb427892a519cab949438d5723d5cc4eb22a09c77644ca7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "e50433b243591e415c959f148bb6f213561204ebc900340790adcb2e48a25182", "generation_seconds_batch": 24.176812999066897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.0146708488464355, "genuineness": 4.507347106933594, "identity": 0.7596758604049683, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que es.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/346c55e1fe5cc2c9dfb427892a519cab949438d5723d5cc4eb22a09c77644ca7.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17c8f46b2d33cae3e2bd80446c20baeeaf644fe440c052d6c6d4c07b137f09dc", "generation_seconds_batch": 16.778370628017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.3984374833747786, "ce": 6.403861999511719, "genuineness": 4.300237655639648, "identity": 0.37028607726097107, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "34c009083af5f2956ade5469c5139e49ad45e67804247e50332a566a5a6ac2ee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "949bbab799de300c29ec1d2c9d78017be7ecc9755e97fd9e54b5f607cf4aa922", "generation_seconds_batch": 21.459020851063542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.133810997009277, "genuineness": 4.3881025314331055, "identity": 0.8274048566818237, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/34c009083af5f2956ade5469c5139e49ad45e67804247e50332a566a5a6ac2ee.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": "34c753d3840afcfa96f680210e324f77e49a935a6bc34ff72a150784182a559f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a642f3367cd551a20452c763722f5cf81fa38c896adc2f7b4b0e00fcaf3adafd", "generation_seconds_batch": 22.190684631001204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 6.38797664642334, "genuineness": 4.243753433227539, "identity": 0.6895158290863037, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/34c753d3840afcfa96f680210e324f77e49a935a6bc34ff72a150784182a559f.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "34ef1b87623fccd6d26e3239e1efd8ed3d1237141083f13ddc35725f8623c8ba", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2b4119a05b38b2c554ecd7154f25289e09ceab9805ead7ef5cd46ed1be7c6c4a", "generation_seconds_batch": 19.721828431007452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.221180438995361, "genuineness": 4.444586753845215, "identity": 0.5062392950057983, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/34ef1b87623fccd6d26e3239e1efd8ed3d1237141083f13ddc35725f8623c8ba.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": "356198d4efee784191030806eb70e5961fed733a96d974537e92e818b8da9656", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "777e2d74e598a65b73b920d78b8b93c583c96c9ca7bf65617b5555eed5644f54", "generation_seconds_batch": 20.95910456997808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.6932271718978882, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/356198d4efee784191030806eb70e5961fed733a96d974537e92e818b8da9656.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": "36c5747460bf253bea1ad3a5026db9a1fdc2f7656b0905b699cbc70dbb6b1a98", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1387e906cfd8be20b002f0984519ef166c10e6e7d37dc21cae194cf1c181d2e", "generation_seconds_batch": 18.326804143027402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.475345611572266, "genuineness": 4.093128204345703, "identity": 0.3936021625995636, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/36c5747460bf253bea1ad3a5026db9a1fdc2f7656b0905b699cbc70dbb6b1a98.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "36f58f860eb12a032786a37c2a619fd56ef1fdbab9f645fa85590ded7636faf5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "836a05d73f0c0f793a6fd574ed40ee983239c29956aa9572e0944b8d94796f50", "generation_seconds_batch": 22.04918524902314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.0312499778330384, "ce": 6.443574905395508, "genuineness": 4.243753433227539, "identity": 0.48742514848709106, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/36f58f860eb12a032786a37c2a619fd56ef1fdbab9f645fa85590ded7636faf5.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "36f9e91f9bc4e6158396a296b711c9d3af6d4d20e6d849f6f014f76d0044245a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f717a7ffe2f0ccb0da99f1a7948a1b5f898299e4cffe096655cd26747afae4ae", "generation_seconds_batch": 28.872232612106018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.086154937744141, "genuineness": 4.042920112609863, "identity": 0.7219358682632446, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/36f9e91f9bc4e6158396a296b711c9d3af6d4d20e6d849f6f014f76d0044245a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "372f678bd1d8edfb32f3b290fccacb60508e851e44ea5736b2534345791e1eeb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "3238426231e56b2e7be88e9177a6fd9d71be341eb059ae031ec2df37af5cdd5d", "generation_seconds_batch": 22.81979105202481, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.8396116495132446, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/372f678bd1d8edfb32f3b290fccacb60508e851e44ea5736b2534345791e1eeb.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "37952c70de89f6548f93566c012db92476a42f1c5e9af5c816ee4468369abb45", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "fce068e360feb7cc806f265621bf76d816a2481556e618cfc7da4b313d1c8532", "generation_seconds_batch": 16.91690432606265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1203866004943848, "target_z": 1.8359376459324985, "ce": 6.205295085906982, "genuineness": 4.362998008728027, "identity": 0.7186306118965149, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/37952c70de89f6548f93566c012db92476a42f1c5e9af5c816ee4468369abb45.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "37e6f8aa89fbfb1f19b38812aadd367c10279fb5eea33bb4ce8bbb90118fb78f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "845225789ea1d93001d0e0fafdc6ad59e89742c99687505e78ebdcf01f55cffc", "generation_seconds_batch": 21.1096476779785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.292664527893066, "genuineness": 4.432034492492676, "identity": 0.7563961744308472, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/37e6f8aa89fbfb1f19b38812aadd367c10279fb5eea33bb4ce8bbb90118fb78f.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "37f54eb8614ae46cbbeaa37885e60510b10891f473797b0514ad0bde8abe25a6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4f9fdaf24f789e3cf6fa69f91cc332149fa41dd53d5058fc2f9b93b20c055065", "generation_seconds_batch": 22.55567723698914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.14062495936057, "ce": 6.332377910614014, "genuineness": 4.093128204345703, "identity": 0.44908958673477173, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/37f54eb8614ae46cbbeaa37885e60510b10891f473797b0514ad0bde8abe25a6.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "391308d7a5be318f046c18b5451c901f3a064622f1a20486b3f8a3ee7f743e5a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "58d9d817be769cdc9d9fddf916293b922dc88fa46232edee4e35dd5b675d9848", "generation_seconds_batch": 21.832369793904945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3011136054992676, "target_z": 1.0234375129307278, "ce": 6.252951145172119, "genuineness": 4.425758361816406, "identity": 0.7767196893692017, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/391308d7a5be318f046c18b5451c901f3a064622f1a20486b3f8a3ee7f743e5a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "3992504c0438b8a89b333b81b8be81ea3659ec995e8c8ad5892729fe57d82cf1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "8853ff678e2161cd7101f2170202dc71a45cbef3a4abd7b908de76c4d2f79014", "generation_seconds_batch": 22.34909993398469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.062326908111572, "genuineness": 4.193545341491699, "identity": 0.7669436931610107, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3992504c0438b8a89b333b81b8be81ea3659ec995e8c8ad5892729fe57d82cf1.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "39927630a496f776b3d737870a5b5e6e8650ddd9de5e7b1375a9528d9629f915", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "02ed8957fc9bd4ddd4b727145aa9eb31b20894332de0c93770b4fe09fee7e632", "generation_seconds_batch": 21.761320462916046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 5.990842819213867, "genuineness": 4.105680465698242, "identity": 0.7732124924659729, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/39927630a496f776b3d737870a5b5e6e8650ddd9de5e7b1375a9528d9629f915.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "39eb6f385f06830447b285022d9c7c1c94516a585eb97632a6ab65d67d6bca21", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "81abc5358f5b06f3c7afebf81c86e482aa63c33559046579bf1bf87730c770c8", "generation_seconds_batch": 21.766492787050083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.5888870358467102, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/39eb6f385f06830447b285022d9c7c1c94516a585eb97632a6ab65d67d6bca21.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "39fa5475f3671aefd8a52b27902505a24f24bc4d0f7e7637dec6f6e848e7cdef", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c257faab8e6e806f9a55ab6ff2d0951e3bd47e657658ac28deedc031d50aa69d", "generation_seconds_batch": 22.635253492975608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.16719388961792, "target_z": 0.8906249002486719, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.6453331708908081, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/39fa5475f3671aefd8a52b27902505a24f24bc4d0f7e7637dec6f6e848e7cdef.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "3a0a9947fb6d7e7f08c9c127bc1b10e18302887f91fcd0dd4734cd1b751f14cb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "14e925c558eafc186e0100bf26b00fe5247461f3e0fae5c509bfdc115ef8bf72", "generation_seconds_batch": 16.374133631004952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2696032524108887, "target_z": 0.9921876533214857, "ce": 5.808161735534668, "genuineness": 4.532451152801514, "identity": 0.6706347465515137, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3a0a9947fb6d7e7f08c9c127bc1b10e18302887f91fcd0dd4734cd1b751f14cb.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "3a435e62ec43b72930d839d7431d2d97da4f1792176bb2896234b2c5bfbfc88e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "af4f0cf732b5eb60eebcd9400fe8780511350c53f176c26f79fb77daa82c6d2a", "generation_seconds_batch": 16.552627812023275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 1.820312479680285, "ce": 6.197352409362793, "genuineness": 4.425758361816406, "identity": 0.6600729823112488, "hypothesis": "Ik heb het klein doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3a435e62ec43b72930d839d7431d2d97da4f1792176bb2896234b2c5bfbfc88e.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "3a9161e23989b379db14113bf5149e29337dda8f6c29704640a164511856d526", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "329cadb01664e00bc3d854e6eb1168579ee65fa8c3a44ac64d775f520762ba95", "generation_seconds_batch": 15.928961253957823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.173524379730225, "genuineness": 4.582659721374512, "identity": 0.7308467626571655, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3a9161e23989b379db14113bf5149e29337dda8f6c29704640a164511856d526.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "3aa16039bc6fe4b7e8fc0a93f60bc3a2826344c38fc41571f29343e8f52ef521", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "13ef9993cb234040b28d18dcde8c618cfbc91380c0c8e941554a37581923255f", "generation_seconds_batch": 21.077634089044295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 1.0468751440852517, "ce": 6.316492557525635, "genuineness": 4.369274139404297, "identity": 0.7494760751724243, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3aa16039bc6fe4b7e8fc0a93f60bc3a2826344c38fc41571f29343e8f52ef521.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "3ab81717ec4dbeaacef58bd238d4a764c439850b95565fbb6011f146368f7682", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "033c18b3fa41900c13d20d1203237ab7b00cfc740c1179353222c4c87c969b19", "generation_seconds_batch": 15.876779804006219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9938862323760986, "target_z": 0.7187499630550637, "ce": 5.982900142669678, "genuineness": 4.2751336097717285, "identity": 0.5873898267745972, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3ab81717ec4dbeaacef58bd238d4a764c439850b95565fbb6011f146368f7682.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "3afd3210d39f0212b1006f7ed20dbc4bb66fa179d65bcc8c039d4ff3752142ed", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9312bb1e6bbec7375164226ae5fb57e5748eed1c33f68f428b2620121ac40c3a", "generation_seconds_batch": 16.835587586974725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8520890474319458, "target_z": 0.5781250036944936, "ce": 6.395919322967529, "genuineness": 4.545003414154053, "identity": 0.4655340313911438, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3afd3210d39f0212b1006f7ed20dbc4bb66fa179d65bcc8c039d4ff3752142ed.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "3b522c316811267da72496db34693fdde3eade4ae61c4b106b6e47c59cb1ec9f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "dc7591de910282c1be4f4401af8c05673babcd9bbc6cb34ba348ed2d5799ff1d", "generation_seconds_batch": 22.16830753395334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.1640623540675015, "ce": 6.300607204437256, "genuineness": 4.13078498840332, "identity": 0.3048174977302551, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3b522c316811267da72496db34693fdde3eade4ae61c4b106b6e47c59cb1ec9f.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": "3bf8d5f064c9b12c46ff3fb5739ff03507d6b79d4c7ed5cc71ec67f6e71dc874", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b70951354cb612b6d169f6ff56d2f1dbbd35bc08ea538467f199c014b2751442", "generation_seconds_batch": 16.611532305018045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.324435234069824, "genuineness": 4.319066047668457, "identity": 0.6906455159187317, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3bf8d5f064c9b12c46ff3fb5739ff03507d6b79d4c7ed5cc71ec67f6e71dc874.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": "3c13b4c4fe0bc815a25d22703d8178bc6a45eed43bee6bc02bc6a7bfed143c95", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5efb6edbbf99b15e43e5129d3a27c72f4e9b5e861a19d60c7bde5f41ac7c10a9", "generation_seconds_batch": 22.77096488000825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.038498878479004, "genuineness": 4.293961524963379, "identity": 0.6941388249397278, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3c13b4c4fe0bc815a25d22703d8178bc6a45eed43bee6bc02bc6a7bfed143c95.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8893213224e299268b2aa72ce0658e217246953f99a6a0bcadcefcb57c70e51c", "generation_seconds_batch": 15.6802391349338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.7029755711555481, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "3db4959fbf53efbe0e3d23c3927b92002c5e86ebc3bedecd47bb7e398b234c90", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "37321c49e64b3e56c32214769af7f42a564a0014229a9ea6591321c176b1db4c", "generation_seconds_batch": 22.638845414971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 5.998785495758057, "genuineness": 4.482243061065674, "identity": 0.7060260772705078, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3db4959fbf53efbe0e3d23c3927b92002c5e86ebc3bedecd47bb7e398b234c90.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "3dcb04e9f4b36a2a13aec1103e56c7e9980c0be53ccd71c0cd994c272a36dd5a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "5f648928655e71a4427f0d83eb945b4f1276f6554f57e94165f6bcf2cd58df30", "generation_seconds_batch": 22.777148399036378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.260893821716309, "genuineness": 3.9299514293670654, "identity": 0.23681595921516418, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3dcb04e9f4b36a2a13aec1103e56c7e9980c0be53ccd71c0cd994c272a36dd5a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "3e48d5e91bc98e2235f914273da2d5a3f210fd8f56053bdc40589ab0bb9c2d99", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "901514994829cfac29f1e061bf48107e2f5e6e6bf1846336837e297fbf398b28", "generation_seconds_batch": 22.758334049023688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.1640623540675015, "ce": 6.1020402908325195, "genuineness": 4.532451152801514, "identity": 0.723619818687439, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3e48d5e91bc98e2235f914273da2d5a3f210fd8f56053bdc40589ab0bb9c2d99.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": "3e6ef6d83d74818eed367293a37eb7d6deba2cb58e1e436ab5738caae39aca7c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2333097d8dc8ba8d55431bda3fa73376cfbb7698f88a86ac5d9075e49a1f43d4", "generation_seconds_batch": 15.692302871961147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.931323528289795, "target_z": 1.6484375424866768, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.7317687273025513, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3e6ef6d83d74818eed367293a37eb7d6deba2cb58e1e436ab5738caae39aca7c.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": "3ebf4fc9d17b99feb17cefb69a73346c0d918e2ea475b7d76c58aaa61c9bafe5", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a88e93e40a82992d1809bbc6b5b764379f84ebdccf0e6e6fd8f4a8f80e3b4f41", "generation_seconds_batch": 17.286179161979817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.7760178446769714, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3ebf4fc9d17b99feb17cefb69a73346c0d918e2ea475b7d76c58aaa61c9bafe5.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": "3ee5487b6790a9be111aee49132d020facf7f12a0fd7294078f544120225700d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2473ca9389f131efb785db7a0887567c77d7912e6f60fbd566cf2a2f656e25b", "generation_seconds_batch": 21.724717332050204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.062326908111572, "genuineness": 4.325342178344727, "identity": 0.658099889755249, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3ee5487b6790a9be111aee49132d020facf7f12a0fd7294078f544120225700d.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "3f2cd31f29247bfd4ec5783c83df53398f3dde841e4a766381238f6a58cb2301", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "36127083820586b78e4e387587e83287aab0c8a123a1bac92e5e906f59ba0d51", "generation_seconds_batch": 25.411646236898378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.109982967376709, "genuineness": 4.469690799713135, "identity": 0.5688436627388, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3f2cd31f29247bfd4ec5783c83df53398f3dde841e4a766381238f6a58cb2301.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "3febd73ee3bb7fae36b022a6f9b44ceb1583c302653b720a70e0e780b9d61bd1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3819f26d73bb4d34169b91111b0d592aaeefe94a671fc0fa2f13135c0acec50f", "generation_seconds_batch": 23.061784959048964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 5.186647415161133, "genuineness": 4.193545341491699, "identity": 0.8132274746894836, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/3febd73ee3bb7fae36b022a6f9b44ceb1583c302653b720a70e0e780b9d61bd1.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "40140ff9933abd12269ebec7b046b689c9e60e38e21094a62ef2efe6dbe9521a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "dac17d63e41dd23d445c37f41ed29d994dab2dfca0abe5b190c9130bd31d237b", "generation_seconds_batch": 16.079922795062885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.260893821716309, "genuineness": 4.030367851257324, "identity": 0.25128090381622314, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/40140ff9933abd12269ebec7b046b689c9e60e38e21094a62ef2efe6dbe9521a.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "403116cc5f6c3493261d1e34fc25151a75d2b3859b0e4f57cb4503d2902dcebf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "04dbe7c515f481ee55d50bcb3c4d742022088fe5da6a4d2355f5852980196af9", "generation_seconds_batch": 16.51503476803191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.3515624575133232, "ce": 6.205295085906982, "genuineness": 4.419482707977295, "identity": 0.7205295562744141, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/403116cc5f6c3493261d1e34fc25151a75d2b3859b0e4f57cb4503d2902dcebf.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "40492ebc9c093d3eb31fd6da643ba9174218e8df441be33e900e95a34f511078", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "defc57bc042ed08528f76c4cecc861966e8f3dd5d23afa3d12e50fa21b59add2", "generation_seconds_batch": 20.66833991999738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.340320587158203, "genuineness": 4.193545341491699, "identity": 0.6646486520767212, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/40492ebc9c093d3eb31fd6da643ba9174218e8df441be33e900e95a34f511078.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "404f04cbfe36fba4849f85bfc6b609134c2d05619e2ae3152b718552bd366fad", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4f7687c0d3640c5f818ab99b81a8471bc6d379aabbc1c2dae99ae7a399a023e7", "generation_seconds_batch": 15.085791699937545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 1.179687520319715, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.7756123542785645, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/404f04cbfe36fba4849f85bfc6b609134c2d05619e2ae3152b718552bd366fad.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": "406b828a6b814a9378c7532036aed7085193d198d127d4391bff2b758900d042", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45cd74f2098124af8f270164baddb0c028421d85c947290979d1c4c4252404f3", "generation_seconds_batch": 16.856845609028824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.2890625018472468, "ce": 6.372091293334961, "genuineness": 4.356721878051758, "identity": 0.5779733657836914, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/406b828a6b814a9378c7532036aed7085193d198d127d4391bff2b758900d042.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "4082c0bea0473ab0dc50130aeb584426437d81384dbd09d131179bf3f6001179", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "da32f4f143969a56bc519aad73074d6c4afa4fd400fcebe7272e8c1c62401b6e", "generation_seconds_batch": 21.483887915033847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5566781759262085, "target_z": 0.28515626939609157, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.5152180790901184, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4082c0bea0473ab0dc50130aeb584426437d81384dbd09d131179bf3f6001179.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "40d56873106caebbe4e7b55ffe35a3188bd4b5d6df87da93aeccf239e0a8feb4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2f0d6f14869a5d0cc65ed2b8e72a97b59c4adec3d98149d8eafd271cc2415717", "generation_seconds_batch": 15.085791699937545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.0146708488464355, "genuineness": 4.469690799713135, "identity": 0.690937340259552, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/40d56873106caebbe4e7b55ffe35a3188bd4b5d6df87da93aeccf239e0a8feb4.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "40f2c95183c6059260ff24f793bcda051d82088fa539b482ea7c7aab8e88261a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "17fa3b2d00a0f2757096ba86486fd7acf92cc50261b237df63cf9d0490d14590", "generation_seconds_batch": 21.857127454015426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.109982967376709, "genuineness": 4.005263805389404, "identity": 0.2780584692955017, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/40f2c95183c6059260ff24f793bcda051d82088fa539b482ea7c7aab8e88261a.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "40fa2250ea8dadc6843409f83ebe2e6dd5b1d772cff3ee83f91da785ec420593", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ad32e5f64d7cb1a194f3c1401f008db12bca4f08d7da43ddcb43c45d4dab7fae", "generation_seconds_batch": 22.04918524902314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.234375011083481, "ce": 6.4515180587768555, "genuineness": 4.18726921081543, "identity": 0.5977155566215515, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/40fa2250ea8dadc6843409f83ebe2e6dd5b1d772cff3ee83f91da785ec420593.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "419de444f2c0ddc2a7b89cd4f7de746ff859585545c30c621bed60c752fa9f34", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1892dda49f855b3a9028299164d1bb39509ff1e6a09f83be563276b9061dc56f", "generation_seconds_batch": 22.334150686045177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.6054667234420776, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de nuevo.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/419de444f2c0ddc2a7b89cd4f7de746ff859585545c30c621bed60c752fa9f34.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "41c311d5bec431637ecbb5c04f0194d0062e73d5896fab9816e093a9eecd74e8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "14eb3b8fd0a47d3fa310618d7a02207b67a593cc215ced51dbd129218382a981", "generation_seconds_batch": 16.05603813799098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.100234270095825, "target_z": 0.8242188303552365, "ce": 6.173524379730225, "genuineness": 4.162164688110352, "identity": 0.48690086603164673, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/41c311d5bec431637ecbb5c04f0194d0062e73d5896fab9816e093a9eecd74e8.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "41d5228ad45c40667a37b919cdea30b3d69e06ff7f43d4c62ca3cd446c96cd7a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "06d7234ca9d0678ef992ec1b611bf7f46d8d485d896db9c3c790385920a7a031", "generation_seconds_batch": 22.673301325994544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.6789664626121521, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/41d5228ad45c40667a37b919cdea30b3d69e06ff7f43d4c62ca3cd446c96cd7a.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "41fe875d5cd88c3e4ec2080a8f8e1fb5c3ebd26b2bc31d8a17b71f9025c83023", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8bbd89d3764d6b0e8593fb2c21d5a157fe8a48bc34b5f88ba5481675038afe42", "generation_seconds_batch": 20.62663569196593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.308549880981445, "genuineness": 4.137060642242432, "identity": 0.09848606586456299, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/41fe875d5cd88c3e4ec2080a8f8e1fb5c3ebd26b2bc31d8a17b71f9025c83023.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "42200c4cba45c685c519a1abdb16242de035f1a3a9abb8b4e9e311cca21bf718", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "599bc1a2634afedd73612e799f6a0c7f4052502fbd844bf704927a2ed06b7312", "generation_seconds_batch": 22.105579582974315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.226276159286499, "target_z": 0.9492187416873893, "ce": 6.213237762451172, "genuineness": 4.494794845581055, "identity": 0.651533842086792, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/42200c4cba45c685c519a1abdb16242de035f1a3a9abb8b4e9e311cca21bf718.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "428a1a77cccbe630eb37f860f907c497d7a3f172b3f519d04805cac9f04dc5de", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "42a6334b6be5e2c253df373978d58ec6470cc73038a333f0cd356747f4e2160c", "generation_seconds_batch": 22.667715633055195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.459460258483887, "genuineness": 4.6956281661987305, "identity": 0.4371981918811798, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/428a1a77cccbe630eb37f860f907c497d7a3f172b3f519d04805cac9f04dc5de.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "42a58e6b06ee9f80b3d65713393139e967bd1ab494287bf929c3e48381d9a7eb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6abbe952338b9637123b8e2f4ec74141504c5d40c51a46d90e9e405c6b8ca216", "generation_seconds_batch": 20.318221117951907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2696032524108887, "target_z": 0.9921876533214857, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.7750520706176758, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/42a58e6b06ee9f80b3d65713393139e967bd1ab494287bf929c3e48381d9a7eb.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": "42f7b7336318074ae2a9bba5ebe87a932a157731b0b57b0b106f3954dbe527af", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5bd041d7dfe668c9a9bbd8d426fbfd13cb31f51634cc6aea114720ff02c78bd", "generation_seconds_batch": 21.0285669779405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.403861999511719, "genuineness": 3.998987913131714, "identity": 0.30684036016464233, "hypothesis": "I wrote the number on the envelope, read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/42f7b7336318074ae2a9bba5ebe87a932a157731b0b57b0b106f3954dbe527af.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "430f1043afac1b5452c815b633a6899c6411df361c84fd9586bf662c2a545765", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c30753b1f846e562b9889dd294a17290359a7ff43cebd60dab013178f0e1a8a2", "generation_seconds_batch": 21.628867999999784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1829493045806885, "target_z": 0.9062500665008854, "ce": 6.133810997009277, "genuineness": 4.356721878051758, "identity": 0.6356889009475708, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/430f1043afac1b5452c815b633a6899c6411df361c84fd9586bf662c2a545765.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "432f12701c77272b8a5e675bdc181b73858954f2ebc611cc9d043a8b18253801", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "5d10c36e3a5350c308bc5ce9adb7006a98ccfa93aa1787faf77fac0723cbb352", "generation_seconds_batch": 19.94164912297856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.600463390350342, "target_z": 1.3203125979040813, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.6900433301925659, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/432f12701c77272b8a5e675bdc181b73858954f2ebc611cc9d043a8b18253801.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f9d9a778cd9d9b2391bc6a2c55d38c8e73fb15b7e1a6fa7fe976b9b1c5f7f0f", "generation_seconds_batch": 18.72672589798458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.9999998817762038, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.7995911240577698, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.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": "43623c514010fcd1e31399c652cd74abbfa69b251e763ba46ec635181ed8e59c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dee8dfdbdbbe790143abe1148f8d7d2a79e5a358567771b9e0ee4175005e264a", "generation_seconds_batch": 15.845448528998531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 6.356205940246582, "genuineness": 4.13078498840332, "identity": 0.33813053369522095, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43623c514010fcd1e31399c652cd74abbfa69b251e763ba46ec635181ed8e59c.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "43838e6a640cd00a02255268a64a41bef0ba57ddf808e1d1a0607bf60729bdc0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "ac255003e8feff00430355e2f64b56b7384fd0bc1d6bd7caf43670141735fc5b", "generation_seconds_batch": 22.168450308032334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.165581703186035, "genuineness": 4.268857479095459, "identity": 0.7743953466415405, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43838e6a640cd00a02255268a64a41bef0ba57ddf808e1d1a0607bf60729bdc0.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "4399bc78f10b32ad988396c23ba70063964f6ee7fa4993c6392153a66ef2cefa", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "8b5f0f029e7a11d7e40f85de4d5976b42735993a88c3280230a5783581d503e8", "generation_seconds_batch": 14.575893678003922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.835937409484906, "ce": 6.062326908111572, "genuineness": 4.331617832183838, "identity": 0.5853034257888794, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4399bc78f10b32ad988396c23ba70063964f6ee7fa4993c6392153a66ef2cefa.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "43a347d6aef1e708443562fed3afc31b1587d674bff5d329c2f87961887fa501", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bab440e3a4cc4dfcd1f28ef1422272d2db0e89726c8275b6e28b3d24bcd6eac", "generation_seconds_batch": 21.360278938082047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.054384231567383, "genuineness": 4.2249250411987305, "identity": 0.7566943764686584, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43a347d6aef1e708443562fed3afc31b1587d674bff5d329c2f87961887fa501.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "43c3367dc279404dee9fe4d93f510ddfa424a66d3c624a1149350bc51472765b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "62bf32581d7408a07b45dbdf7bf7f5c255d019b7f50561e82be153066872add2", "generation_seconds_batch": 21.422150991042145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.104172945022583, "target_z": 0.8281249445825956, "ce": 5.974957466125488, "genuineness": 4.369274139404297, "identity": 0.749970018863678, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43c3367dc279404dee9fe4d93f510ddfa424a66d3c624a1149350bc51472765b.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": "43d7395596becdca94348045c492f489f629df2979a8d90de43bcddfcf5e2a8d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbe0e563f6acece627b659c862f06c71b916474e6517c99f3790009b2b04b9f7", "generation_seconds_batch": 20.55520050600171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.203125151474239, "ce": 6.252951145172119, "genuineness": 4.362998008728027, "identity": 0.6354141235351562, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43d7395596becdca94348045c492f489f629df2979a8d90de43bcddfcf5e2a8d.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cf6dd0e1f38313067188b4de7a010c2a60b6eb5f9e59d799308f2115091f1c3", "generation_seconds_batch": 15.388435151078738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.09409761428833, "genuineness": 4.2374773025512695, "identity": 0.47253572940826416, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "44429e539747245712e58dcd2be044074d987dc5296d3857f50549255fbffd6e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e5dc940a99d6cda4d38fd53cbc9b08331b5410d4642c3bbd0cffba8c7d48a853", "generation_seconds_batch": 20.150674605043605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6796874020959187, "ce": 6.229123115539551, "genuineness": 4.162164688110352, "identity": 0.5513620376586914, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/44429e539747245712e58dcd2be044074d987dc5296d3857f50549255fbffd6e.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "444308d821f5399f98ec1f5a9f5af5c0219b895872f86f0df7d3981b5f62ac4f", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5449245a0b1c29f33e1b83826aa78821415314f4a7e1535205902ac36606e6d8", "generation_seconds_batch": 16.866492770961486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.131744623184204, "target_z": 0.8554686899644784, "ce": 6.395919322967529, "genuineness": 4.362998008728027, "identity": 0.17473968863487244, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/444308d821f5399f98ec1f5a9f5af5c0219b895872f86f0df7d3981b5f62ac4f.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "4499ad58caa9c0354b45f5ef07d377faf509129184acc45234cda004d81bae52", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6f28d98307eca4f5bf94f05f3769f774e034f4c98f403513694a4f8b084177ad", "generation_seconds_batch": 16.764418622944504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 1.1484374242628805, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.29647961258888245, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4499ad58caa9c0354b45f5ef07d377faf509129184acc45234cda004d81bae52.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "44a854a5884d21c914e8c8dd4968d8f32d1f710c96fe3a37c025193e81ae4e87", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a470d21257835c0dd869ca87c28aa693202bc7d47feaca9c606e44d1dcbcde04", "generation_seconds_batch": 17.107057873974554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.7421875942095877, "ce": 6.0305562019348145, "genuineness": 4.18099308013916, "identity": 0.574956476688385, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/44a854a5884d21c914e8c8dd4968d8f32d1f710c96fe3a37c025193e81ae4e87.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": "44d1b98ec428307f6a39f461cab42632a307d17308291675f2847e02a71a9cea", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2abfa271533b39d66f84507e1064975444111ffe0c0c32403b624eb241f5f0a7", "generation_seconds_batch": 22.600771839963272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.395919322967529, "genuineness": 3.898571252822876, "identity": 0.6062208414077759, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/44d1b98ec428307f6a39f461cab42632a307d17308291675f2847e02a71a9cea.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": "4520713adc2296fba9c2105ac6904f032768c847f9fe1aadc1153b3844115a40", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3d8a4077f48454425bd128c2c4a43e54e11ef62b0fcd548b7161614965fcc24", "generation_seconds_batch": 23.622165640001185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2656643390655518, "target_z": 0.9882813026465342, "ce": 5.95112943649292, "genuineness": 4.7834930419921875, "identity": 0.6077858209609985, "hypothesis": "J'ai le numero sur l'enveloppe. Lise le bien and verify the two dernier chiffre. Appelle moi quand you're a moment.", "wer": 0.5454545454545454, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4520713adc2296fba9c2105ac6904f032768c847f9fe1aadc1153b3844115a40.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "453fabcf94d1ee54a91d7dbca64e46e02ed0143c339fa1ea0123cca95d70fbb5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37325efe3cbb78e46dcc489617fd730af605d1e0077f3dcbb6558209f0bce7da", "generation_seconds_batch": 22.291195678990334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.149696350097656, "genuineness": 3.923675537109375, "identity": 0.24488113820552826, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/453fabcf94d1ee54a91d7dbca64e46e02ed0143c339fa1ea0123cca95d70fbb5.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c586524aa97646bb1b2260880578283ee0216126ecc22d8d37670d800220f522", "generation_seconds_batch": 15.957885914016515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.3593749224156335, "ce": 6.252951145172119, "genuineness": 4.017816066741943, "identity": 0.578757643699646, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "47ac9f45214d0ca9f4a51f5451c6b0e3539e7dafdecc6a1c4256084afa3464e4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e12926106ae218ef05d4766a7a887dfd844707ee55c4d2d4d9be2fff683796f0", "generation_seconds_batch": 21.33407437708229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.2187498448312675, "ce": 6.348263263702393, "genuineness": 4.3127899169921875, "identity": 0.49067193269729614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/47ac9f45214d0ca9f4a51f5451c6b0e3539e7dafdecc6a1c4256084afa3464e4.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "448299006dba92a4567c14ae3c2830d0a9a60982f748279c9bda89b0e665f595", "generation_seconds_batch": 16.56279383704532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6796874020959187, "ce": 6.078212261199951, "genuineness": 4.344170093536377, "identity": 0.719894528388977, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "486a7184c2fea0cf147c933ad9d8140090b6c48131050183f23db6fb61d84bbb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "7a2e1282215f5ff75c2b5f0cd720ba2e197103f42cc61cad6fbab213080cb0f4", "generation_seconds_batch": 19.208777489955537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.09409761428833, "genuineness": 4.105680465698242, "identity": 0.7412165999412537, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/486a7184c2fea0cf147c933ad9d8140090b6c48131050183f23db6fb61d84bbb.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "488014c7717ca1b86f74cfcae844a7c62268b8d0c5ec218aa9e8f676d831a673", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d082a2dde32db3074e84c74e2ed018efc514345f68dfd3842083ad7a76bdf205", "generation_seconds_batch": 16.46867455204483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.117925643920898, "genuineness": 4.319066047668457, "identity": 0.6925128698348999, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/488014c7717ca1b86f74cfcae844a7c62268b8d0c5ec218aa9e8f676d831a673.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": "494599a1276e1d62ea29e7f8247df35bd2e41f8a0ed2c9b588e3ffdc898b243b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59f9b1dbca384fa3244cb227e2f9575644905575e4d57f1467ee7b7ba01de105", "generation_seconds_batch": 16.54500030190684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.1874999852220254, "ce": 6.022613525390625, "genuineness": 4.482243061065674, "identity": 0.6652765274047852, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/494599a1276e1d62ea29e7f8247df35bd2e41f8a0ed2c9b588e3ffdc898b243b.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "49b6bff418a9b31250ea013a0a262cdea0e85ea9f09ffccfdf0f9a7c687b0979", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b52d2804c0fc1ee77c908a43979275a48c4511bc15c93457ad052aa565b18e3a", "generation_seconds_batch": 17.28891885804478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.078212261199951, "genuineness": 4.080576419830322, "identity": 0.4916861951351166, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/49b6bff418a9b31250ea013a0a262cdea0e85ea9f09ffccfdf0f9a7c687b0979.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca83f55ecfd891d3bb61628b4e1d842ce17ade9720ca0587a946d41c0c6262c1", "generation_seconds_batch": 14.77064359805081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.7384201288223267, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "4a5b6d85714eee0a3c1a2434bd4af932ed71c3f78697a0238f9af03df7ab53b7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5e54f97c7ed1e2c9e4c96b3d9fa96126a1ac0b155c7825611501ad832277af07", "generation_seconds_batch": 21.1096476779785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.24500846862793, "genuineness": 4.243753433227539, "identity": 0.7322851419448853, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4a5b6d85714eee0a3c1a2434bd4af932ed71c3f78697a0238f9af03df7ab53b7.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "4aaf7207e1c0c849e4329f5fdddf735129c2379f98a9597ba83d2becfbe87457", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1a93d18f90bd305c03954ad942b5f78302890eae52a8055723320fb53880d303", "generation_seconds_batch": 22.73400146397762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.24500846862793, "genuineness": 4.306513786315918, "identity": 0.8208826184272766, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4aaf7207e1c0c849e4329f5fdddf735129c2379f98a9597ba83d2becfbe87457.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "4ba4a318490ac5876a8fd658b07939e4ae7a10fed39363dc0386db0616a53fe4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "78864ed6911148bd9b827413002a9c953de148cd32461f7093f5589b2d0596cf", "generation_seconds_batch": 22.0146220800234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 5.839932441711426, "genuineness": 4.030367851257324, "identity": 0.559925377368927, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4ba4a318490ac5876a8fd658b07939e4ae7a10fed39363dc0386db0616a53fe4.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "4bf7204e483662d5ecbfe5ebb380d9cea07b145b02cccc2602ae4ee96162e996", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d83ef431be3db951fb4dc34dd8ffbc579ed94f1f1fa93dcce6d28a43c3076293", "generation_seconds_batch": 21.37317533302121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9032936096191406, "target_z": 0.6289062620071043, "ce": 6.141753673553467, "genuineness": 4.720732688903809, "identity": 0.6589040160179138, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4bf7204e483662d5ecbfe5ebb380d9cea07b145b02cccc2602ae4ee96162e996.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": "4c21473886ca0e7a1093041b37c96d02f1686a18e3f3922297ad8f4c7d73d5a0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa21fae93bdc550d4ba516a7a00bd150e25bd420e6f1ec059bea5a625eaffd52", "generation_seconds_batch": 22.780558011028916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.2187498448312675, "ce": 6.006728172302246, "genuineness": 4.519899368286133, "identity": 0.7715651392936707, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4c21473886ca0e7a1093041b37c96d02f1686a18e3f3922297ad8f4c7d73d5a0.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": "4c2781594f65d8b01bb6dfab21b7ca6f26b1db4b9af814fbb9bf63c1b0a7bf2a", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62e048c79c7dda8eb3ba226ffb6c73a82eed1b27ba71fcf8afb3fc16091698fb", "generation_seconds_batch": 14.404132487019524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.7109374981527532, "ce": 6.054384231567383, "genuineness": 4.331617832183838, "identity": 0.6185096502304077, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4c2781594f65d8b01bb6dfab21b7ca6f26b1db4b9af814fbb9bf63c1b0a7bf2a.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": "4c9ab26a32c61d372f65b902a4eda41201410d5ee602ae5d52c0b1e2d5e0359c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b2212b558aa1cdcdfb9ad6e302a5658c78dbb37ea625810598dadc4a3c66668", "generation_seconds_batch": 16.72495257493574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.5566115379333496, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4c9ab26a32c61d372f65b902a4eda41201410d5ee602ae5d52c0b1e2d5e0359c.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "4d10d413b6ebcd705e6dc554226519d53f69d1dc02d7b4c71cf7e75afd63fd69", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70bf1db333436c4956e24ba2d34483081c2847424d5cdafb01489fb01f01cdf5", "generation_seconds_batch": 22.168450308032334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.093750169946707, "ce": 6.086154937744141, "genuineness": 4.055472373962402, "identity": 0.7135603427886963, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4d10d413b6ebcd705e6dc554226519d53f69d1dc02d7b4c71cf7e75afd63fd69.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "4e21b8b28f50cbd07d4c768e5aa111bcf0c30b69bbdc76121bb0d217b9f97edf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b5a1354f56acb18c070175275bba7ef155fb5c13e83d424c956544d11a573c52", "generation_seconds_batch": 16.572594539029524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.09409761428833, "genuineness": 4.13078498840332, "identity": 0.685326099395752, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4e21b8b28f50cbd07d4c768e5aa111bcf0c30b69bbdc76121bb0d217b9f97edf.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "4e431b1e9a8ed22da0a237ae2587b5eae5f62b3074e60c62a6b8804d4b41cd73", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "94cf8f3ae7a4b8cca6634a115b1fdff38346ba4aaa3ca88b6e7f979600f707e8", "generation_seconds_batch": 23.135817307047546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.332377910614014, "genuineness": 4.425758361816406, "identity": 0.6030794382095337, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4e431b1e9a8ed22da0a237ae2587b5eae5f62b3074e60c62a6b8804d4b41cd73.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "4e8b4959aea33f0e771e2e39c2fc2ede1d2343124883a2d64035065ef3b3da7b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5715c2e5129949c34dc771a62cc02e3010b8af99370d9d0ada316c9d30bdc874", "generation_seconds_batch": 20.449848075048067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.181467056274414, "genuineness": 4.413206577301025, "identity": 0.525875985622406, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4e8b4959aea33f0e771e2e39c2fc2ede1d2343124883a2d64035065ef3b3da7b.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "4f2cff96726127bcf710debfae0f15abae956db5abd25ea15acfdcc3b7855c82", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad8b05010a2650675467597f6506b76f9854ba2f846689da641df5fadc6b7e4a", "generation_seconds_batch": 22.55567723698914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726963520050049, "target_z": 2.4375000443339236, "ce": 6.41180419921875, "genuineness": 4.105680465698242, "identity": 0.5250238180160522, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4f2cff96726127bcf710debfae0f15abae956db5abd25ea15acfdcc3b7855c82.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "4fc9b3d37dc5513a2111a3e3e26f5583bb34ca2be26839f19f17ae6a333b8c27", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 131, "seconds_generated": 10.48, "mp3_sha256": "e1f99ac64dea47fcab6a1d8f218c70f372da96b248483de04229835f168720d1", "generation_seconds_batch": 16.374133631004952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4609374390408552, "ce": 6.09409761428833, "genuineness": 4.6203155517578125, "identity": 0.6725131273269653, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.28000000000000114, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/4fc9b3d37dc5513a2111a3e3e26f5583bb34ca2be26839f19f17ae6a333b8c27.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "5056c284973cbe72b7dec12f25ebe504ae52d54f36109dcfdab74ce3302c6a48", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5d51affaa6b3fcdba324315744222ddf4f05ace41f5f97c982a9e08dac3de6a5", "generation_seconds_batch": 23.385027041076683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135683536529541, "target_z": 0.8593750406394299, "ce": 6.0305562019348145, "genuineness": 4.557555198669434, "identity": 0.6168372631072998, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5056c284973cbe72b7dec12f25ebe504ae52d54f36109dcfdab74ce3302c6a48.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "50b128fa96cba16c2ef254faf7805bd839b27088d5e36eee27381b4a7391a676", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1712c701db81f8eff975f3671e5b65e63ade26faa3b736c0d233bf24cf3981fb", "generation_seconds_batch": 25.377015238045715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.5150015354156494, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/50b128fa96cba16c2ef254faf7805bd839b27088d5e36eee27381b4a7391a676.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": "51d93c5b50431b4b42816cff36151b6b0f1891d493914c4d173d821cd7ac28e9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a76d3db54a10c13ae815dbd647e6119e4efeddb04ad7afbe96fcd5a8ea417c1", "generation_seconds_batch": 16.31743995693978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.149696350097656, "genuineness": 4.507347106933594, "identity": 0.7241036891937256, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/51d93c5b50431b4b42816cff36151b6b0f1891d493914c4d173d821cd7ac28e9.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "52bf8ce79145745da713e81a0c28eb0aa0d8f7326c8c891a406c7697be714f18", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1d36a4a3978a566362136fdce05b583960b16f98ff2dae118ea8a763ad3cfb50", "generation_seconds_batch": 22.758334049023688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.09409761428833, "genuineness": 4.319066047668457, "identity": 0.8146711587905884, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/52bf8ce79145745da713e81a0c28eb0aa0d8f7326c8c891a406c7697be714f18.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "52eb447d78cdcf6b1556c8b117be338373b886ce191f2fbc311b55a03d13c1d1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dcbf2ef1e88769251f4c791be8c7125192d38d83ad83b164c32640ca7484ed94", "generation_seconds_batch": 22.24954691703897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.281249800497344, "ce": 6.395919322967529, "genuineness": 4.13078498840332, "identity": 0.5774537324905396, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/52eb447d78cdcf6b1556c8b117be338373b886ce191f2fbc311b55a03d13c1d1.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "5307002f46187366a12d43d457fd9a363327ca494a9f9223f7c9b70af30c2152", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ae52c09204e88e78ded48f67733dcd6012156a583aa6663ae8f2bb06664773ac", "generation_seconds_batch": 23.88473197107669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 5.80021858215332, "genuineness": 4.469690799713135, "identity": 0.8121945858001709, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5307002f46187366a12d43d457fd9a363327ca494a9f9223f7c9b70af30c2152.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "532c516db7081b0c202ebe2cc5bc370f53dbe8a57ffb1ecabee31cf1b55aa285", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b67a937619695315e667dadde938d72cb2deb29bc16171d0962aca414d5d36b4", "generation_seconds_batch": 22.973399496055208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.171874818969812, "ce": 5.967014789581299, "genuineness": 4.042920112609863, "identity": 0.5497961044311523, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/532c516db7081b0c202ebe2cc5bc370f53dbe8a57ffb1ecabee31cf1b55aa285.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "536401e7242a70f3e7cedbae0ed07add2a00b2328bdecbe97fba1c2d74725135", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c5e696fdeacb89c1f8a416910a559e2d490afd7c400bb1d4ca6bd94228ec1977", "generation_seconds_batch": 23.21977522096131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1278059482574463, "target_z": 0.8515625757371195, "ce": 6.117925643920898, "genuineness": 4.6705241203308105, "identity": 0.6349009275436401, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/536401e7242a70f3e7cedbae0ed07add2a00b2328bdecbe97fba1c2d74725135.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "542e2106ba82615a72c6848f72502e4f27780d499330d9f311a23eb044ace60d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b6047464d725af750b2474e242501837bb0aaa269af0e0b171009fb3fc0fbe61", "generation_seconds_batch": 22.272128722979687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 5.967014789581299, "genuineness": 4.068024158477783, "identity": 0.6049251556396484, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/542e2106ba82615a72c6848f72502e4f27780d499330d9f311a23eb044ace60d.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86db65105846af0c47f4a5ab44a0e6478b5a6b5b3a45a7ee1025b8072dc91f9b", "generation_seconds_batch": 15.957885914016515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.300607204437256, "genuineness": 4.017816066741943, "identity": 0.5090315341949463, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "54af199a59746d2afcb3403d0073a9b7e9a478401009381b3d8f461bb0b4e63a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3b6387d05ed913071d31f28ded83580a6da0f5ec0aafac72f136d5573dca2df9", "generation_seconds_batch": 22.698530990979634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.9609375572646512, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.6591088771820068, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/54af199a59746d2afcb3403d0073a9b7e9a478401009381b3d8f461bb0b4e63a.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "54b0f39fdb783e01771dc408fb7ebecd641b27306a1d0aff1ed3bd806e2cdc84", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dd090458838c85c23bd0b3cbedbedd163fd3874d6c8fa2238e108e639afedab2", "generation_seconds_batch": 21.590355616994202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6481871604919434, "target_z": 2.3593749224156335, "ce": 6.284721851348877, "genuineness": 3.9550557136535645, "identity": 0.5393298864364624, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/54b0f39fdb783e01771dc408fb7ebecd641b27306a1d0aff1ed3bd806e2cdc84.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "54cfd6f24adb51d480c3b8e602a5a22f34de30890389bcd6a99b24c0a141b74a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9703e5ca1f8bdd04e0f2c83d07a9d71985b355a38d5acc172538a5b59641cc57", "generation_seconds_batch": 22.76982721593231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0214579105377197, "target_z": 0.7460937084369467, "ce": 6.483288764953613, "genuineness": 4.469690799713135, "identity": 0.30754104256629944, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/54cfd6f24adb51d480c3b8e602a5a22f34de30890389bcd6a99b24c0a141b74a.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": "55a098d5dd66499c670464f1d002f5a5df619a1e8711c925937268f364d2bb48", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fb19f69551336f5a07df55e54794580b639505e27c3e4bbfd12cbf20ca3604d6", "generation_seconds_batch": 13.53319508396089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3405017852783203, "target_z": 1.0625000738898727, "ce": 6.09409761428833, "genuineness": 4.193545341491699, "identity": 0.6885616183280945, "hypothesis": "Je bericht kwam nadat ik vertrokken was, ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/55a098d5dd66499c670464f1d002f5a5df619a1e8711c925937268f364d2bb48.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "55a7718332df2dfb6263c05a265f56dccea2416f92aeebb007648c11c1383758", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "73e569c80c8405d608a2f1e13623a3fdcc4ccfc5e15c03ff34f6fde082cb397d", "generation_seconds_batch": 22.24954691703897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.41180419921875, "genuineness": 4.168440818786621, "identity": 0.4987855553627014, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/55a7718332df2dfb6263c05a265f56dccea2416f92aeebb007648c11c1383758.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": "55f811b468c797e894d8f99889e1438ecc79131fe7abbf8a48be8aa12b793db0", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ba02fe2eda51c0b2bb9b65d58a77b11a01bdeb92910ad971b6d2b34b43bf452", "generation_seconds_batch": 20.71616707299836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.38003396987915, "genuineness": 4.099404335021973, "identity": 0.27667897939682007, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/55f811b468c797e894d8f99889e1438ecc79131fe7abbf8a48be8aa12b793db0.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "561453c381c0d8ed47343bc6412abb0179205dc671c4668a34f2dc2778d4e83a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "aad21b4560746c5f214a4a8ab5deee6c8b32423754ac032b9faf8e061707153c", "generation_seconds_batch": 21.968828371027485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 6.316492557525635, "genuineness": 4.369274139404297, "identity": 0.5659369230270386, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/561453c381c0d8ed47343bc6412abb0179205dc671c4668a34f2dc2778d4e83a.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "56360c71ca6a9c248f41b431d065155d4f3355c9c0466e836a246d2ecc958f79", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "81ce328349c8c041eb9a03eb86735677958d0326c05c326b520eda0a15c55f2b", "generation_seconds_batch": 21.95830267400015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135683536529541, "target_z": 0.8593750406394299, "ce": 5.982900142669678, "genuineness": 4.557555198669434, "identity": 0.7300662398338318, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/56360c71ca6a9c248f41b431d065155d4f3355c9c0466e836a246d2ecc958f79.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "5687bd787976340f8923885fd209ed1d1cff997a0ca25c01668ea5472f28a551", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 126, "seconds_generated": 10.08, "mp3_sha256": "749838638dcf7426b031fd1504ff9dab8ea618dda37a620fcc5d246e7f425750", "generation_seconds_batch": 23.66228427691385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.078212261199951, "genuineness": 4.469690799713135, "identity": 0.7431268692016602, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble.", "wer": 0.15789473684210525, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5687bd787976340f8923885fd209ed1d1cff997a0ca25c01668ea5472f28a551.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "56a45764d4f0603bb6e88a4ef4bf74727ac990ed0dbeda79b116531f5f6a437a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9f94fe3506e7c6101c0505f7b840ddf205092afef5e8ea98575d86d7730b46d5", "generation_seconds_batch": 21.674366755993105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.149696350097656, "genuineness": 3.810706615447998, "identity": 0.3670007586479187, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/56a45764d4f0603bb6e88a4ef4bf74727ac990ed0dbeda79b116531f5f6a437a.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "570c64717f613218d81156d70fe15d641b99adedb2599a8cbdcadbcb206ffa1f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f4a1560400e57f137bc5e00d17f1af32044fd710e6d3b5e2d48384fe74d9406f", "generation_seconds_batch": 21.616909770062193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.3671873873179443, "ce": 6.070269584655762, "genuineness": 4.400654315948486, "identity": 0.7375035285949707, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/570c64717f613218d81156d70fe15d641b99adedb2599a8cbdcadbcb206ffa1f.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "5789042a0756e26f9c835f30a3cd4804d285e98337efff68764f27afda739024", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "115bc99a2e409a89683ea424736db5b6761512ceeeff9eb315e48b1e32c1bced", "generation_seconds_batch": 22.18294645601418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.109982967376709, "genuineness": 4.162164688110352, "identity": 0.5474929809570312, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5789042a0756e26f9c835f30a3cd4804d285e98337efff68764f27afda739024.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": "57f6a033d9f7ccfcd788240815f605928eb30efe294a81cc60ab030db8b90013", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5825112976de962bdd581501ab896ab4475ff4565c3ef4752f5ef82f7e4f5cc2", "generation_seconds_batch": 14.609708734904416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2656643390655518, "target_z": 0.9882813026465342, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.8847050070762634, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/57f6a033d9f7ccfcd788240815f605928eb30efe294a81cc60ab030db8b90013.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "5820a9f8d71cf11680b63cbee7d14448f8be043f3f3e66caa5124db9e28d711e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e7f42422454172365ac8b48e1c3bd626c8a56f2308e2069a704e0001ffc1162", "generation_seconds_batch": 21.73912115301937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.046441555023193, "genuineness": 4.231201171875, "identity": 0.8490661978721619, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5820a9f8d71cf11680b63cbee7d14448f8be043f3f3e66caa5124db9e28d711e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "584b92f60e7da8ac6488981934e4338928bd584cee8a0c5f1dd13fce5aca6c37", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0378681bf18a7e16f8758a6ee96beb519cc2e2c09bf6f3fc7144316a70d8f909", "generation_seconds_batch": 16.09147557907272, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.704218327999115, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/584b92f60e7da8ac6488981934e4338928bd584cee8a0c5f1dd13fce5aca6c37.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": "589104f34c81563f1efd2cdb955182c60ac45669114abfdcd9167a364312ac1f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cea37e62e8d27314c685b01170bddaf340b8a20d79f0248f624e1ffcde18f1a7", "generation_seconds_batch": 14.059106701985002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 1.9296874612078168, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.8052800297737122, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/589104f34c81563f1efd2cdb955182c60ac45669114abfdcd9167a364312ac1f.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "58ecdc2afc0457e962bf0956daa9e7f889bda3913b7aa1c504dbe4a2ac24c78d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "94053fa16c8591cc426ce49a42c89540c6960c139ab73548559f11f22db31770", "generation_seconds_batch": 16.21689886599779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.25384783744812, "target_z": 0.9765624870692723, "ce": 6.181467056274414, "genuineness": 4.15588903427124, "identity": 0.3101046681404114, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/58ecdc2afc0457e962bf0956daa9e7f889bda3913b7aa1c504dbe4a2ac24c78d.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "59fc32de48e2d2e181b035477b9d63782815fc1bad7099ca578ca1d0431fddc3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7e3c0b29f75af4e13eb693cb208899eb59289153e7dce45edf8b46d2257ed948", "generation_seconds_batch": 16.329043647972867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.419747352600098, "genuineness": 4.243753433227539, "identity": 0.5283614993095398, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/59fc32de48e2d2e181b035477b9d63782815fc1bad7099ca578ca1d0431fddc3.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "5a005f238e98101616e22548e92563cd9c852b78e1c4c898f3ca3505ecc75bd6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "82c16a1a79ef86ee736010989159d0823071d7098223a890eb75536f64236633", "generation_seconds_batch": 17.70522850099951, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.348263263702393, "genuineness": 4.250029563903809, "identity": 0.6417364478111267, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5a005f238e98101616e22548e92563cd9c852b78e1c4c898f3ca3505ecc75bd6.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "5a1191e1e54b89e180ad89138077d3d825a4788ec439e87e5d4e7cc98893adc5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4e624c7cd62b8154aee90a262d5ea024b5ed023060468ac80f79a01126d4f2ec", "generation_seconds_batch": 21.023523634066805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.348263263702393, "genuineness": 4.469690799713135, "identity": 0.5968115329742432, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5a1191e1e54b89e180ad89138077d3d825a4788ec439e87e5d4e7cc98893adc5.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "5a509e6288001b2275431af182baa50b1d7111df5be30dea07d604faf96b9b93", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0008e3356e93620b3dabf6d74d79b2abaa63263c41abba2966fe6b3dcb2481ef", "generation_seconds_batch": 21.497531311935745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.046441555023193, "genuineness": 4.231201171875, "identity": 0.7698667645454407, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5a509e6288001b2275431af182baa50b1d7111df5be30dea07d604faf96b9b93.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "5a9121790b0da8a5d8119823beaad3ee2e89bb0fd4ee938a4fa9ff4691fad355", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6ed4a96511bc1e61012a327ab01152e713dc5de6aa9e4ecb61c127b6787f23f4", "generation_seconds_batch": 16.20112233608961, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 6.348263263702393, "genuineness": 4.356721878051758, "identity": 0.530414342880249, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5a9121790b0da8a5d8119823beaad3ee2e89bb0fd4ee938a4fa9ff4691fad355.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "5aa04c055a88787441f7b5f9fc2fea1f616e3f6aed35fc06cea9e2e6c557dd76", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "dfce77eb9f1ed60aafe13a93cc81344bac32d96c1b91619b122c8aba6e7348ef", "generation_seconds_batch": 16.518168410053477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.419747352600098, "genuineness": 4.444586753845215, "identity": 0.3347848057746887, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5aa04c055a88787441f7b5f9fc2fea1f616e3f6aed35fc06cea9e2e6c557dd76.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "5ad27b0c9773640e8f74d18ad40fe97e15110882fb64568f27995b6e5c4b53ea", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7cc8caed4c0c70bcceee0cd165895346b109454de687cf6e73502dbaa382f90b", "generation_seconds_batch": 16.10289249906782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.292664527893066, "genuineness": 4.519899368286133, "identity": 0.6470924615859985, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5ad27b0c9773640e8f74d18ad40fe97e15110882fb64568f27995b6e5c4b53ea.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "5c06ead6b3a0c06b15e7f5e439eaab3157813e3c296644f4790d3cbe40f8d7ed", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "0d38c51545694ffa6477de1a6153b7e2192510be6f89635e093725c109d11017", "generation_seconds_batch": 16.654883901937865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9374999261101273, "ce": 6.09409761428833, "genuineness": 3.898571252822876, "identity": 0.677384614944458, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5c06ead6b3a0c06b15e7f5e439eaab3157813e3c296644f4790d3cbe40f8d7ed.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "5c7fc4ddaf1b56feeb00de2049f8c3856776ebd675ed8f0fe1ff35d053ab655e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "8debab39d32157d61c6bb1128dde091a0eca09b21f01d1cf71723296d5812f0e", "generation_seconds_batch": 26.066171929007396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.372091293334961, "genuineness": 4.419482707977295, "identity": 0.6242411136627197, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt, Quanksus et seul Caire. Dis-moi quand tu es prêt.", "wer": 0.5555555555555556, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5c7fc4ddaf1b56feeb00de2049f8c3856776ebd675ed8f0fe1ff35d053ab655e.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": "5ccbd31f441be49752cb4a319b706942c498cce9aad32487d8cf1333ffe97abf", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27b41f0a024472b6e8da59487d78ec0f9da476302edcf7fd406246481d5ec51d", "generation_seconds_batch": 16.109124653972685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.205295085906982, "genuineness": 4.557555198669434, "identity": 0.7153885364532471, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5ccbd31f441be49752cb4a319b706942c498cce9aad32487d8cf1333ffe97abf.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "5ced503a0df6423b593a2fd5d11d7aa069f58fde406d5f8e5a2263b69c5d813a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "77f0421bd1948d7e0984a6bd52cf719cef7e536fda0ba6c83694c5617043620e", "generation_seconds_batch": 20.449848075048067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.133810997009277, "genuineness": 4.457139015197754, "identity": 0.8297970294952393, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5ced503a0df6423b593a2fd5d11d7aa069f58fde406d5f8e5a2263b69c5d813a.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "5d88478785575186ee1df17df53dda6570387e5621717c43707d90cd00a58445", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "818eef94ff79b93a14cf3ef9bac9405e3d452de8a7477f990ce6a0ac536c44e3", "generation_seconds_batch": 15.689369329018518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.5048171877861023, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5d88478785575186ee1df17df53dda6570387e5621717c43707d90cd00a58445.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "5d8ba103551700638af1057cc6071dbee66a0f603ce8ce9ac9e97a327da3e2cf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ed4e5687731a888c6d648d18d09b9aa84b76cd2fc2f3ea36f4727a128adf8970", "generation_seconds_batch": 22.565222267061472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.863905429840088, "target_z": 0.5898437010479594, "ce": 6.268836498260498, "genuineness": 4.846253395080566, "identity": 0.7926788330078125, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5d8ba103551700638af1057cc6071dbee66a0f603ce8ce9ac9e97a327da3e2cf.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "029d8bdae49d803ff949c87315c8aea025a1e3d817d443ae1aa8744e5e37022a", "generation_seconds_batch": 16.970804449054413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 1.1328124944582596, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.8189260959625244, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "5e5e72adfa90fd417f5c8c3fce2c20e68b7235a9808ffabeb87f3eab8a66cc50", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8fedd4f426c2795cb7352232dc171dcc3c4df1225a55c6fd444637f8231b5c45", "generation_seconds_batch": 21.71362113195937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.348379373550415, "target_z": 1.0703125387921832, "ce": 6.427689552307129, "genuineness": 4.507347106933594, "identity": 0.3642764687538147, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5e5e72adfa90fd417f5c8c3fce2c20e68b7235a9808ffabeb87f3eab8a66cc50.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "5eeca915d5a4277ff9e741e3c6a1512bb3883d652a1ba7dc471ccbac13065941", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d30335a2d0103cec6f6b0002d14f6fe2a6327ad309149814783b557a8c4cc4ad", "generation_seconds_batch": 21.775620901957154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.895416021347046, "target_z": 0.6210937971047938, "ce": 6.372091293334961, "genuineness": 4.645420074462891, "identity": 0.3379116356372833, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5eeca915d5a4277ff9e741e3c6a1512bb3883d652a1ba7dc471ccbac13065941.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": "5ef4a63209cf9f42e3721753bae7325a948b693c580cee4c2f44a2539e3027ab", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbf0eac04e8b92e4e91129271fe61c40d387741c1e193000a35b0a31a1c56c14", "generation_seconds_batch": 15.265560391941108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.125868320465088, "genuineness": 4.231201171875, "identity": 0.6420022249221802, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5ef4a63209cf9f42e3721753bae7325a948b693c580cee4c2f44a2539e3027ab.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": "5f33ff0adb8c57a88bf4d69859e78c3e4ff4e5e53109c9cb4c3f553ec8b5ac6b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cd140c122169562dc018bacae2f21791f7f1f738461f9c6e35ccb6b36c89c42", "generation_seconds_batch": 13.668674846063368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.2890625018472468, "ce": 6.24500846862793, "genuineness": 4.570107460021973, "identity": 0.705358624458313, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5f33ff0adb8c57a88bf4d69859e78c3e4ff4e5e53109c9cb4c3f553ec8b5ac6b.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "5f44b91664a3853e345dfacefbfee86bb7c9dc3891e03c8ad0585e4f363449d1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "71193842e61e32a50bff08bffd5ae7e4f1e6cfb647193d5e3ae55d7484792365", "generation_seconds_batch": 15.580986023065634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 5.903473377227783, "genuineness": 4.111956596374512, "identity": 0.616051435470581, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5f44b91664a3853e345dfacefbfee86bb7c9dc3891e03c8ad0585e4f363449d1.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "5f63deb21db43a1a38fbccc51293d9278298e832614e414278deb51d944a368c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6bc7aa08869bc618b5d2bb32dcf707f06d033e57aaa8b7fb9f6b755008945404", "generation_seconds_batch": 22.71272316505201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2420313358306885, "target_z": 0.9648436714920103, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.28794366121292114, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5f63deb21db43a1a38fbccc51293d9278298e832614e414278deb51d944a368c.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "5f99f0f78dbb51142cb2852547a6371892bef4e5995d933991cb1612f3ba00eb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9b6b2d7bb952f3fa39605a4e7a0f54d329cdc1832f99a7c5ecac1c96ef91d38d", "generation_seconds_batch": 28.09001506504137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.9999998817762038, "ce": 6.006728172302246, "genuineness": 4.494794845581055, "identity": 0.623985767364502, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5f99f0f78dbb51142cb2852547a6371892bef4e5995d933991cb1612f3ba00eb.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "5fbd653c6396f752a7617599db14b8cd4c57adbe4edd3b646824f980e9c235af", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "961d9ed6107b66a1d8927effc8bbfffc92805126ec6d3b58434c3d3ff8516a47", "generation_seconds_batch": 22.98339196201414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.4687501403907581, "ce": 6.197352409362793, "genuineness": 4.168440818786621, "identity": 0.3371521234512329, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/5fbd653c6396f752a7617599db14b8cd4c57adbe4edd3b646824f980e9c235af.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "60923b473a99741c6508b72d75f6e8f7a1614bf628bd7f51b5e313095aae6ee0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "fb366ac4122138a9d18887062f5a89ad83c254b2f90e6a1a7f0dd17bf571fb4c", "generation_seconds_batch": 15.773359431070276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.221180438995361, "genuineness": 4.030367851257324, "identity": 0.6243733167648315, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/60923b473a99741c6508b72d75f6e8f7a1614bf628bd7f51b5e313095aae6ee0.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "60ddb65b155b9ae970293ecdbe084f1cd805d59dfa78b01fba13164ead525a73", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2fb89677f7f3dc831f1ca8c946f2076723b236bc305c61306ca81d4e181f1bb4", "generation_seconds_batch": 22.969700744957663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6319737434387207, "target_z": 1.3515624575133232, "ce": 6.205295085906982, "genuineness": 4.250029563903809, "identity": 0.6746382713317871, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/60ddb65b155b9ae970293ecdbe084f1cd805d59dfa78b01fba13164ead525a73.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "60e7d4a956fb04fb4e4f91bd5139bb9aea6f4b2d94d503229d44b5a70384f1ea", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "fe6c3b7d161832ee754e91ad932b83a73c33782e09119ee59bce7d5534617702", "generation_seconds_batch": 17.28891885804478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.086154937744141, "genuineness": 4.300237655639648, "identity": 0.8147372007369995, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/60e7d4a956fb04fb4e4f91bd5139bb9aea6f4b2d94d503229d44b5a70384f1ea.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "60eb0787c64560a42a3576698f6e5c60b61f1a652ea490d7dc145aab0adac70f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "467b8fff4643e14334b11da7c985dd539b8e7551eacde7bd300e35129198d6ca", "generation_seconds_batch": 22.248808316071518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.6403909921646118, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/60eb0787c64560a42a3576698f6e5c60b61f1a652ea490d7dc145aab0adac70f.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "6147accd96be091df5183500dbfcad822fa6c05148f220ac1b7af59c15131fac", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4f7f73afb09d863be7bff9ef68e9d29e3b90a50d5ef4d995ee24e38402386c47", "generation_seconds_batch": 15.133732657996006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.149696350097656, "genuineness": 4.20609712600708, "identity": 0.7107470035552979, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6147accd96be091df5183500dbfcad822fa6c05148f220ac1b7af59c15131fac.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "6149e6ee0d1f01644fa92afbe0c2a7daa7dadc244ad391aab39f5d4ef35aa69f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f739f36b30d0ac0ad2118af2aaeb0cbd93934419f0f8e845087780989073823b", "generation_seconds_batch": 22.514748272951692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.348379373550415, "target_z": 1.0703125387921832, "ce": 6.324435234069824, "genuineness": 4.268857479095459, "identity": 0.19976598024368286, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6149e6ee0d1f01644fa92afbe0c2a7daa7dadc244ad391aab39f5d4ef35aa69f.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "619dfab7db509705b7f947bf8acd908e4bc65a21f2e191366bd0ff9c000153b7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "3d0a01677fd9235e2b080ce9438d0780ff3c0119efbd043e1935faa5fd0adaa7", "generation_seconds_batch": 19.208777489955537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 6.181467056274414, "genuineness": 3.9801597595214844, "identity": 0.643164873123169, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/619dfab7db509705b7f947bf8acd908e4bc65a21f2e191366bd0ff9c000153b7.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "61c789a7564659966e1cf6ed815dd5083bf05ff094ceeb757a216be1c39cadd0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0a60ca1267ae361a8fd92d1eca3c72d59b0638a71635e31e5dd2a17809ea4992", "generation_seconds_batch": 21.755624048993923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.078212261199951, "genuineness": 4.143336772918701, "identity": 0.80840003490448, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/61c789a7564659966e1cf6ed815dd5083bf05ff094ceeb757a216be1c39cadd0.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "61f1d23d5d2dc7f6150c25de70662922c8909e2862bd1bcbec1c23e8e6a2c97e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "872c1188dc96ae6812a3f4cbd6c86b254b5905e6b78d5c1b5fbd01d3895cc7c0", "generation_seconds_batch": 21.960687801009044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.117925643920898, "genuineness": 4.15588903427124, "identity": 0.8040785789489746, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/61f1d23d5d2dc7f6150c25de70662922c8909e2862bd1bcbec1c23e8e6a2c97e.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": "62564a420d289e8a598ed73b8a1775f5f25d110780e82bdb86cb25928488c8ef", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f619394704e099d8a7494eb5a35063b772ddf51aab332f60b336bc98cbafd63e", "generation_seconds_batch": 15.783159850980155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.173524379730225, "genuineness": 4.344170093536377, "identity": 0.8265643119812012, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/62564a420d289e8a598ed73b8a1775f5f25d110780e82bdb86cb25928488c8ef.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "627dbeaab31f6e04e641469dff877f806fb2a9e3573ba691a3d3a66accde14f4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "448624afe7d39098248fef11a4a6a9460f7f0f831aea671957988e58257ccec8", "generation_seconds_batch": 22.66937733395025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.316492557525635, "genuineness": 4.469690799713135, "identity": 0.21651607751846313, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/627dbeaab31f6e04e641469dff877f806fb2a9e3573ba691a3d3a66accde14f4.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "62bf689f57da189424c2aade75172a9a671affb149ef2d58c4b7e390fc286b06", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ab31459a53d97d2f9e037154b4a56e56c73bc4acccf1a49a1dc62cb6a98a7f1c", "generation_seconds_batch": 21.194156382000074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9741921424865723, "target_z": 0.6992186825754912, "ce": 6.38797664642334, "genuineness": 4.300237655639648, "identity": 0.2854187488555908, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/62bf689f57da189424c2aade75172a9a671affb149ef2d58c4b7e390fc286b06.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "62c1774794154a4bc9c177e06a1e2b0b0823bc4418cbe22e94f31ab6ab3dee01", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c9400dfe3f3d902a1047403472f3929025cf5f66b58547d235b935240394eb13", "generation_seconds_batch": 26.671142132021487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.6091376543045044, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/62c1774794154a4bc9c177e06a1e2b0b0823bc4418cbe22e94f31ab6ab3dee01.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "62c564ee63027d30023606d426af00ba53c741fa2a91030e1606a2fea42495f1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7b0230b12f3120798db21ecb16693e699b7cc92f23c8a10ca12d886258eaec7a", "generation_seconds_batch": 21.26672684098594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9155683517456055, "target_z": 1.6328126126820557, "ce": 6.173524379730225, "genuineness": 3.603597402572632, "identity": 0.3274637758731842, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/62c564ee63027d30023606d426af00ba53c741fa2a91030e1606a2fea42495f1.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "62c9521a86e27b1dbbed7417e3778e29c93c88561175a6f949ee988493a9f5d2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "52518b61bfba3592038be98dc3be028989d155a46ca12567bfd9d7efd631bd78", "generation_seconds_batch": 15.929894828936085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.149696350097656, "genuineness": 4.331617832183838, "identity": 0.7292606234550476, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/62c9521a86e27b1dbbed7417e3778e29c93c88561175a6f949ee988493a9f5d2.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "630bf6ae98fe528be653446b8ebbbfe6411c1503d4dc0ad9360304c10e2c32ce", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc77ecc210310c503d669fd8837f28e5ba66075ab138a0648f044af93d51c04d", "generation_seconds_batch": 16.313538299989887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.6142397522926331, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/630bf6ae98fe528be653446b8ebbbfe6411c1503d4dc0ad9360304c10e2c32ce.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "644a541d2c4fef64a6e31c11e53893b5b9996b8050a0157daf2785b83b52ad20", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c452a48238a153f6dfc263b459b2d9867536c4e61b26fd8261b6f9190602d478", "generation_seconds_batch": 20.3644430070417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.252951145172119, "genuineness": 3.911123275756836, "identity": 0.4175126254558563, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/644a541d2c4fef64a6e31c11e53893b5b9996b8050a0157daf2785b83b52ad20.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": "645600786b0202e03952e1fed05968eeae65c2e1d51b6395bfe0511a5ecb1941", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8edb2eb1e0f4189ebb8fabe29974ff49025971d312ae9c7bdd6f6a0a0b551b25", "generation_seconds_batch": 14.059106701985002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868302345275879, "target_z": 1.5859373503730079, "ce": 6.109982967376709, "genuineness": 4.243753433227539, "identity": 0.6940401792526245, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/645600786b0202e03952e1fed05968eeae65c2e1d51b6395bfe0511a5ecb1941.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "64a2257c741684a8eb042e19173de300abf5557429995a9c0fc1b706724d5133", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ba2b6310de8b2f08fe7c4011bee43daa65cfed6fa0374a4dc848822b462397f6", "generation_seconds_batch": 22.67743274802342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9899474382400513, "target_z": 0.7148437306039084, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.8733392953872681, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/64a2257c741684a8eb042e19173de300abf5557429995a9c0fc1b706724d5133.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "64b444089551bdd3cf9830b39fc7a8ec1114ad5c43b7872907190fcb961e59ff", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f5218ef9e6f03593b33c228a803b39b451b6a7566cd8165a8130f2ad93e3563", "generation_seconds_batch": 20.72027807997074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.7849267721176147, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/64b444089551bdd3cf9830b39fc7a8ec1114ad5c43b7872907190fcb961e59ff.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "65237c32fddbb888b9291049ae4c1f98a58561d1873112c7f716965766a9ad5a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3e390ce31022939703a7af0bcddebdbc6f89d4e971548d769459850b1d8dbf3b", "generation_seconds_batch": 21.65600947092753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.229123115539551, "genuineness": 4.369274139404297, "identity": 0.7079707384109497, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/65237c32fddbb888b9291049ae4c1f98a58561d1873112c7f716965766a9ad5a.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": "65578c3700ff89e231ebe2547e61365e0cbc5f9fb18c59cc907a8e47d3c7d141", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87cb14838b39a20c856081c6ec1d74edfb1d738063b1a7f7702b07a7e70c1083", "generation_seconds_batch": 15.757981001981534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.9140625314031958, "ce": 6.41180419921875, "genuineness": 4.406930446624756, "identity": 0.6234716176986694, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/65578c3700ff89e231ebe2547e61365e0cbc5f9fb18c59cc907a8e47d3c7d141.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "65906437d9c3e2f7a3e2e3cf78a4359857fd5b7016b71c8aab13d5881a8e4420", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "fc00bf4824db19c02b1e16f2f8bdf9f04e3cd9237ac3990dea77d56ad59f43b1", "generation_seconds_batch": 20.473212865996175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 5.959072113037109, "genuineness": 3.9048471450805664, "identity": 0.7315303087234497, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo está.", "wer": 0.13333333333333333, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/65906437d9c3e2f7a3e2e3cf78a4359857fd5b7016b71c8aab13d5881a8e4420.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "65efdbf7cc36264ec35455fcc88686c029d6a7f2e16b4a16790dcf4ac6a7e481", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "377de33332b15958a47fb8ee9ec5238cc3665180a82b2da1ad73a589886bbb41", "generation_seconds_batch": 22.66937733395025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.4020710587501526, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/65efdbf7cc36264ec35455fcc88686c029d6a7f2e16b4a16790dcf4ac6a7e481.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "65fb98b82de24d2f64ad28618575ccd3a9b3e6f78f8963a74586107fe4b5636c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1410c55b0f8edac8895025fbb2f4deeb68bb5cbab90d9785424630ff3e71354e", "generation_seconds_batch": 15.065925286966376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9785890579223633, "target_z": 1.6953123319005396, "ce": 6.229123115539551, "genuineness": 4.394378662109375, "identity": 0.706063985824585, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/65fb98b82de24d2f64ad28618575ccd3a9b3e6f78f8963a74586107fe4b5636c.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "66a37d167101124e9f188492f3e0731abbc82621a29b3ca25267b6e6ba2d770b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "a5dc7fa6a3211127b5ee47794743ebbc171b55187acf3e247dc1642476dab5ed", "generation_seconds_batch": 20.238426917931065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.2767791748046875, "genuineness": 4.65797233581543, "identity": 0.8022311329841614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/66a37d167101124e9f188492f3e0731abbc82621a29b3ca25267b6e6ba2d770b.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "67016c36ec486ab7988f001fad3e76cb7f6250f8a45e393c5301ba65f7139ecb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "82a4fe5d53e88d855491ee7d638cd0f91e5f24d544ef201763fee617e20da9f7", "generation_seconds_batch": 21.90544672694523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.0146708488464355, "genuineness": 3.860914945602417, "identity": 0.6986079216003418, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/67016c36ec486ab7988f001fad3e76cb7f6250f8a45e393c5301ba65f7139ecb.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "67b67d76f3e4d4705a72481771ab24630c7d4ac97c7b7026c15d5df8adb87253", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "50a8e045bd7937149e3fcdce4f5647d9f1480d14fc33ab73dc92642b2df0d117", "generation_seconds_batch": 21.1845044180518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 1.9296874612078168, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.690190315246582, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/67b67d76f3e4d4705a72481771ab24630c7d4ac97c7b7026c15d5df8adb87253.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "67b908b9fd86213e75e94b55372530fb7f8d9355f5d510f4a62edd541aaf0698", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2bf8c36de60af3458429e9d646b1917d7b790d4576f0b2a404352a7e07a0b564", "generation_seconds_batch": 21.616909770062193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 5.959072113037109, "genuineness": 4.093128204345703, "identity": 0.6778656840324402, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás latido.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/67b908b9fd86213e75e94b55372530fb7f8d9355f5d510f4a62edd541aaf0698.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "67d7ec6f8994690374fb3f6737a062c2c2f9f64b2956d9a4a0f0e9e22d8d91af", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f6dec4a86205e5622c8b17335a677f799ebbfc3af937d2aa2d0e7e55cd4bf0b1", "generation_seconds_batch": 16.387265507946722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 5.974957466125488, "genuineness": 4.256305694580078, "identity": 0.7209528684616089, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/67d7ec6f8994690374fb3f6737a062c2c2f9f64b2956d9a4a0f0e9e22d8d91af.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": "686a2cfe8252a51773d494662f3197820852d3d21bc627568680fc72bd4d5ff6", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6c7774883014f59210984cfb4c8c35478f587e268d0a8f41575327867bea512", "generation_seconds_batch": 16.10862778907176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1120505332946777, "target_z": 0.835937409484906, "ce": 6.483288764953613, "genuineness": 4.394378662109375, "identity": 0.3160499930381775, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/686a2cfe8252a51773d494662f3197820852d3d21bc627568680fc72bd4d5ff6.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "68ba652e5b822cc570943766e232fc56daab5219a70967c4f1f197e4233299f1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "53ba76f1198191b76db6311d622b119738e4fbf795e0bad00b66465df7b73220", "generation_seconds_batch": 21.183293274021707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 5.871703147888184, "genuineness": 3.9927117824554443, "identity": 0.6925494074821472, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/68ba652e5b822cc570943766e232fc56daab5219a70967c4f1f197e4233299f1.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "68cbe27b4a9f207580c973d4e854b07d04c9ecdc7ccaafb92e4875e42e918cd2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "807152fb173a5cbc7ff22ff07c11666318904a284f0ffce1e1207851e247dc0e", "generation_seconds_batch": 20.318221117951907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.149696350097656, "genuineness": 4.457139015197754, "identity": 0.7811487913131714, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/68cbe27b4a9f207580c973d4e854b07d04c9ecdc7ccaafb92e4875e42e918cd2.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": "68ec8d0cfb1e5753bd4e8561b34678cb5ac92ede2645462d24549dc9193d593b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eef8b261cd43e39f66c1176a7228d91335a801056dd4e7c8999388e120d7b2af", "generation_seconds_batch": 15.69572977302596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.23706579208374, "genuineness": 3.7918784618377686, "identity": 0.24953234195709229, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/68ec8d0cfb1e5753bd4e8561b34678cb5ac92ede2645462d24549dc9193d593b.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "6905822bf65f50fb31f44dd80f7ca85c619db69060e14ac49e9cde2dc3c43d28", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "28df67d067cd7f42d5e1acf268c58f3ffd1812554065d5d2899628c2767fe48f", "generation_seconds_batch": 22.4542823350057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3956451416015625, "target_z": 1.1171875646536387, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.8435235023498535, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6905822bf65f50fb31f44dd80f7ca85c619db69060e14ac49e9cde2dc3c43d28.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "69ccb7b2e23555986c36e3de7c8ee3b62ff93cb933454b136dbae206d409c5c8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5a594b834625dcc16570c115dc94b22e958fff8006156a1e82dc37e9e6e1b8a1", "generation_seconds_batch": 22.83557626605034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.171874818969812, "ce": 6.38003396987915, "genuineness": 4.300237655639648, "identity": 0.6715990304946899, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/69ccb7b2e23555986c36e3de7c8ee3b62ff93cb933454b136dbae206d409c5c8.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "6a13b3333c24845624c794808e4839e048e24a594f1d799cfb18993c83c43aa1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "66903c422eb308db1cf90db1f05c881a0e4a33fcd35dc9f890166ba12077ca56", "generation_seconds_batch": 20.231525473995134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.324746608734131, "target_z": 1.0468751440852517, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.5532715320587158, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a13b3333c24845624c794808e4839e048e24a594f1d799cfb18993c83c43aa1.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": "6a27410cd36d472193434a85235a80db5fba32d9d93a737c272a8e7fe9afc625", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf328c506a7fb35293e6661770bd157a4bd91bdb27941071bb3513977ed8b869", "generation_seconds_batch": 16.245029855985194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.221180438995361, "genuineness": 4.111956596374512, "identity": 0.717117190361023, "hypothesis": "Las llaves están en la mesa de la cocina. Mire en el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a27410cd36d472193434a85235a80db5fba32d9d93a737c272a8e7fe9afc625.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "6a3755bef6901dc06ff2c1ad6e9d7ebc968b968d3d5199d9d6d77c0f4fa07cab", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "02e38472d94b14707be2e48e8786637a8cf1128b7cb831db661b9f9eba4f8ead", "generation_seconds_batch": 15.876779804006219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.922987699508667, "target_z": 0.6484375424866767, "ce": 6.109982967376709, "genuineness": 4.20609712600708, "identity": 0.7705017328262329, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a3755bef6901dc06ff2c1ad6e9d7ebc968b968d3d5199d9d6d77c0f4fa07cab.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "6a4e677ce66dd362ac8c5639bc05176ddae5fa5ad0f4d98c54d50f8d510d8abd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8ef6b942cef8d39590175e3c6a04a4b04e1c73fa875a1dc14590aa727fff72f2", "generation_seconds_batch": 21.387530538951978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 5.9431867599487305, "genuineness": 4.632867813110352, "identity": 0.8100039958953857, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a4e677ce66dd362ac8c5639bc05176ddae5fa5ad0f4d98c54d50f8d510d8abd.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "376a0049d94b0be6e07eb9e7a0ef5b876d19377d6280dd756dc094d2f12d472c", "generation_seconds_batch": 16.430591524927877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0884177684783936, "target_z": 0.8125000147779745, "ce": 6.348263263702393, "genuineness": 4.319066047668457, "identity": 0.2662501633167267, "hypothesis": "I wrote the number on the envelope, read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "6a82e587f60d7fd34811b28e0b16723d474724e9a5c2d94f17174590e518c67e", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "dbc05602ecd8fe565a8573c3068d1601ad7e31fe4c0b674ab0537d3bff8815da", "generation_seconds_batch": 18.20241677504964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.213237762451172, "genuineness": 4.143336772918701, "identity": 0.34135276079177856, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6a82e587f60d7fd34811b28e0b16723d474724e9a5c2d94f17174590e518c67e.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": "6ab68b05fd8679188ad09bf99a4dc87c9d71221fed7fb5849451e994553b5199", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5e1cc42fdf12886c568fdaddb1e0cf8410ea23b013f538d1ac55a085c798840", "generation_seconds_batch": 18.71628958999645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.3641486167907715, "genuineness": 4.469690799713135, "identity": 0.7115727663040161, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu. Puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6ab68b05fd8679188ad09bf99a4dc87c9d71221fed7fb5849451e994553b5199.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "85bea6c8e398601969c7bb2d6097f56789bd95a675992130e570d7760c6ce2c2", "generation_seconds_batch": 15.388435151078738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.52516770362854, "target_z": 0.2539062915630534, "ce": 6.221180438995361, "genuineness": 4.582659721374512, "identity": 0.7205919623374939, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "6c30c874bd96d8508c0b11e763273fc04f5c8aabd9a4fd49528bb5f74dec7a7c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7827abe4c71ad711521b85eb9bc0c15003af5da81a74e39fe676cb534860a54d", "generation_seconds_batch": 16.162636072025634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.046441555023193, "genuineness": 4.243753433227539, "identity": 0.7176932096481323, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6c30c874bd96d8508c0b11e763273fc04f5c8aabd9a4fd49528bb5f74dec7a7c.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "6c3b8c5bad1aececc454052ae4a811aa44f3d4941b10d6710bbcfc43b5bac8ad", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "0cb939e7790e3bcae0744e4afa26e70a15a3f004d7ea004331f19ae2fb424667", "generation_seconds_batch": 22.28865651297383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.4912309646606445, "genuineness": 4.281409740447998, "identity": 0.6741906404495239, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6c3b8c5bad1aececc454052ae4a811aa44f3d4941b10d6710bbcfc43b5bac8ad.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": "6ca145a84ce88a182c9d08aa75e1f17efb915b9c6a6f91ecede7e56ac0541d29", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdb8c715ec583aa0716e416c8ba66abb083c8d893df6115e36293005cf7a3b4a", "generation_seconds_batch": 14.414755597012118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.23706579208374, "genuineness": 4.13078498840332, "identity": 0.5525349378585815, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6ca145a84ce88a182c9d08aa75e1f17efb915b9c6a6f91ecede7e56ac0541d29.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "6cbd9a2810d54c5ea58230e744471d375fd02e2e80e035c074784ae8cf556a76", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1a294bd0024c7e7eedef13a2524b68e90970e4ef3bf37b44599d7c8a55b80d79", "generation_seconds_batch": 21.987151146982796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.9843749519715828, "ce": 6.109982967376709, "genuineness": 4.413206577301025, "identity": 0.723056972026825, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6cbd9a2810d54c5ea58230e744471d375fd02e2e80e035c074784ae8cf556a76.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "6cc1e0b20bc1050a0aeea17961dc3a51202c810e6b5a4211ce025b897723a7b3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d3957a4d0cb5fb9d7149d5f8f6680546b64dbdaf8a23690663db66bbe3a716da", "generation_seconds_batch": 28.872232612106018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 5.95112943649292, "genuineness": 4.293961524963379, "identity": 0.8011078834533691, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6cc1e0b20bc1050a0aeea17961dc3a51202c810e6b5a4211ce025b897723a7b3.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "6cfc0800d69b291972184223bfacd4d4d8ff828071e2d8a73d76ce8b86e833b2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e58f66a46d2989b4de4e191f74146457a18796f3c424ccebef43cd597df8c98d", "generation_seconds_batch": 21.775620901957154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.2526143193244934, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6cfc0800d69b291972184223bfacd4d4d8ff828071e2d8a73d76ce8b86e833b2.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": "6d0e26b40eb856d335c42a85d55f9616f0304514e9bff7f17423935a0a62b32c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e59811f2b5d761baf2b966710e8cf4120f9f31dc76970cdc0821deca76c547e", "generation_seconds_batch": 16.164566145977005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.054384231567383, "genuineness": 4.337893962860107, "identity": 0.8735668063163757, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6d0e26b40eb856d335c42a85d55f9616f0304514e9bff7f17423935a0a62b32c.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "6d582ee59d044637e86512b2879c7b5028a15af6c36fc34d4db99996ce8e69e9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "367f721537ee2d3ee3e833b374a94bf54127a64b8227f2fb7df8ea27eb01c0b5", "generation_seconds_batch": 23.414000008953735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.070269584655762, "genuineness": 4.030367851257324, "identity": 0.677570641040802, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6d582ee59d044637e86512b2879c7b5028a15af6c36fc34d4db99996ce8e69e9.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "6d6a690bef05b522b8f2f6e6db8a653707e8e6fea4ce167263acadc338f54bf0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "8ab5839b9d6b030a7c61a4f36884460ae6881a5df70cd4b432141729662b65fd", "generation_seconds_batch": 21.428563820081763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.252951145172119, "genuineness": 4.086852550506592, "identity": 0.3448021411895752, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6d6a690bef05b522b8f2f6e6db8a653707e8e6fea4ce167263acadc338f54bf0.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "6da01f9877432e741056d1e23da019db115ece89f4e3d6ee6cea60466f0614b9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "af7882232d2544d9078c6e762a21a87de10675e8336ed34bc18e9a22ccd987c9", "generation_seconds_batch": 24.006448970059864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9062500665008855, "ce": 6.125868320465088, "genuineness": 4.243753433227539, "identity": 0.7650902271270752, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6da01f9877432e741056d1e23da019db115ece89f4e3d6ee6cea60466f0614b9.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": "6dc57c62e7773e9d000a16f67bbc782ee4bee5207f702ec4004e3b24040c12a9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "790b0b44131ac2780b4c76b18751e0846f067b47004db782da6b0e7413a5f50e", "generation_seconds_batch": 14.414755597012118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.312500133001771, "ce": 6.213237762451172, "genuineness": 4.049196243286133, "identity": 0.5712723731994629, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6dc57c62e7773e9d000a16f67bbc782ee4bee5207f702ec4004e3b24040c12a9.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "6dfb48d3417d0608fdb7c9d27526105cbd98ea2faebcbbb880e4043f63448604", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "20fec13599d30e7c51486526882f101d7c72e84fe25575a4856076df687248ae", "generation_seconds_batch": 25.68102716805879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.062326908111572, "genuineness": 4.218649387359619, "identity": 0.7325282096862793, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6dfb48d3417d0608fdb7c9d27526105cbd98ea2faebcbbb880e4043f63448604.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": "6e16e7bea38940c1463721a95daee38dd77f201e5bf4e26ba501de0e5290244b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c416c3aff61bb86e3dd8cf0af34ffa523ebd2e1843159302d30fcc285384035f", "generation_seconds_batch": 16.72495257493574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.7038143277168274, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6e16e7bea38940c1463721a95daee38dd77f201e5bf4e26ba501de0e5290244b.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": "6eac374ae1b853341fdd91f8a571ccddbb519561d3fec5cf81799d7282367e66", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "926f43b9a190c9bd3077dede45940bc9b221d447ab39debe68e187b34fec70a4", "generation_seconds_batch": 16.0540003470378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.199162721633911, "target_z": 1.9140625314031958, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.7509174942970276, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6eac374ae1b853341fdd91f8a571ccddbb519561d3fec5cf81799d7282367e66.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": "6eb203ed7cdcf4cb3a57aff2278a32ed81684860a0411ca557f9053b9947b350", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6625f903677188c29c3dc22d7689270241d2a83befc8668c023fed49a133d4d", "generation_seconds_batch": 23.705865920055658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.308549880981445, "genuineness": 4.3504462242126465, "identity": 0.6736699342727661, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6eb203ed7cdcf4cb3a57aff2278a32ed81684860a0411ca557f9053b9947b350.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "6efbd660f6a0b964028136665cba650152b64726e78c0095361fa45658cc27a7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "613c2deb3b45b4d4675ab8f407f79a0d6dd42e2f012b607e981b467e8e975699", "generation_seconds_batch": 26.671142132021487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.5739337801933289, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6efbd660f6a0b964028136665cba650152b64726e78c0095361fa45658cc27a7.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": "6f9aa729659f70257e7820fb78a34acd19d61954336863fa96c602c0736c36e5", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c14f1b4d212c63d123a92e026844184dbb580880166c2815e63979ffd2235372", "generation_seconds_batch": 21.495447322027758, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7187499630550638, "ce": 6.3641486167907715, "genuineness": 3.7793264389038086, "identity": 0.30519843101501465, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6f9aa729659f70257e7820fb78a34acd19d61954336863fa96c602c0736c36e5.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "6fc6af5c9eedbddf50f94ee72b6b08aef2db13baf24547fe65d6aadbf862cf37", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c31c948ebfadaef01f289937c168d12d84cd4c3000ecb55f873326b59dbbf4de", "generation_seconds_batch": 15.58795362408273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230673313140869, "target_z": 1.9453126274600303, "ce": 6.38797664642334, "genuineness": 4.231201171875, "identity": 0.5735617876052856, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/6fc6af5c9eedbddf50f94ee72b6b08aef2db13baf24547fe65d6aadbf862cf37.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "705c5f54666c19955f439a9ea903b7324324eb3062250fc43fbe5ae1c4fd7a86", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6ffd4f551f0b36fcbca91d93a2390742f3aed6bb87e807684656030da23cfdf7", "generation_seconds_batch": 22.350354934926145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.0305562019348145, "genuineness": 4.250029563903809, "identity": 0.6228073239326477, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/705c5f54666c19955f439a9ea903b7324324eb3062250fc43fbe5ae1c4fd7a86.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "7077ffc500c9c527513782b8501bad699334dd7fd32320f74158b6cc3335164f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4d921ee95d2833f1d24b660d1b873c7e54cdb94dd6511aa4b007b9ee09e5810d", "generation_seconds_batch": 19.94164912297856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.141753673553467, "genuineness": 4.306513786315918, "identity": 0.7540819644927979, "hypothesis": "La porte était encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7077ffc500c9c527513782b8501bad699334dd7fd32320f74158b6cc3335164f.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "70d8bf2686e07bec99104171ae0d24dbd4f000148e58e444ba629c30fcd161ee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6e7dc359e1947a28ea1660998c0bf6c8ebcb7dcf05ef194ecc4033860aa3043b", "generation_seconds_batch": 22.667715633055195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.308549880981445, "genuineness": 4.457139015197754, "identity": 0.32828301191329956, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/70d8bf2686e07bec99104171ae0d24dbd4f000148e58e444ba629c30fcd161ee.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "7101d919849134741b8734427675af6b28cf1433224e4546b1829d641261ebe2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0bea3367fd173906a38693144c3141f9db039dd5848be94b9fb1ec3d835172ab", "generation_seconds_batch": 15.83446621207986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0781250036944936, "ce": 5.927301406860352, "genuineness": 4.570107460021973, "identity": 0.7578548192977905, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7101d919849134741b8734427675af6b28cf1433224e4546b1829d641261ebe2.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "711127dd47d80bbeedb1accca39db1c6934639a3e7928feccc12d072661fb401", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "65dad58299621c85b8cfdf9543e61b21bd5b3c70249434a1910761e5b1c7db03", "generation_seconds_batch": 15.928961253957823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 1.9609375572646512, "ce": 6.09409761428833, "genuineness": 4.250029563903809, "identity": 0.6199065446853638, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/711127dd47d80bbeedb1accca39db1c6934639a3e7928feccc12d072661fb401.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": "711d86c846f19e53195fcd5483f16c9bf7080c2a06798d15b8199af8625d088d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0343036131911e6f329546d1a803cbdddebd45c31e8e040465fa5836c862a18d", "generation_seconds_batch": 16.614388455986045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.427689552307129, "genuineness": 4.105680465698242, "identity": 0.34616607427597046, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/711d86c846f19e53195fcd5483f16c9bf7080c2a06798d15b8199af8625d088d.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "715d101d3335df861468ae38e8e6fa24d626b0067894834d278760a59573b79f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "600fc84c0a49649320577d865beae47aa7d64ad52bd57041d7f919da7241bc0d", "generation_seconds_batch": 22.316030805930495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.7977721691131592, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/715d101d3335df861468ae38e8e6fa24d626b0067894834d278760a59573b79f.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "71a13c65643930894f0890ef369a61f3974e73b71c33ad08a161fc5ceab481fa", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "992e21d032059cc781dc787ae868b1fae3713fbbd37c5db8b6bd80da6a6c76a1", "generation_seconds_batch": 25.073003839934245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.5783199667930603, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/71a13c65643930894f0890ef369a61f3974e73b71c33ad08a161fc5ceab481fa.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "727082f0c85a32a653d471bda3571a51be761a52d665b1756ff410389c3a6bcf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b2bd2c7d988bc0bf63ed298c23ac59ccad1905a68150167adf7a6997f27d9311", "generation_seconds_batch": 16.34311248292215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.292664527893066, "genuineness": 4.507347106933594, "identity": 0.8474557399749756, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/727082f0c85a32a653d471bda3571a51be761a52d665b1756ff410389c3a6bcf.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "733429ae039fff1cf2d1096dd3eb3307803ad36e96e347bc854f8f633d09723a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5f7f71f3acda60c78d8be8354290aca74095e06bc4c34bd3711e9917629d1ef3", "generation_seconds_batch": 19.420162008027546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.222337245941162, "target_z": 0.9453123910124378, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.44292449951171875, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/733429ae039fff1cf2d1096dd3eb3307803ad36e96e347bc854f8f633d09723a.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "73768c72886c0419ef213cd9d21ff0ef53aeaf07dd854df2f47cabd3fd465880", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "dd5e1fd0401d2bc5a0e63d47a0b155716db53ad7eb03521a7522156e4210501c", "generation_seconds_batch": 21.483887915033847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.1894097328186035, "genuineness": 4.3127899169921875, "identity": 0.6716312170028687, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/73768c72886c0419ef213cd9d21ff0ef53aeaf07dd854df2f47cabd3fd465880.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": "73b56de0b6353db0bd5025fd60a64bea6a7beb61b85459eb7aee8fe261ef2815", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5665b785b8f34c9af820a1c439129a1532afa503a800659b13f5ecde05c977d", "generation_seconds_batch": 16.164566145977005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.6592882871627808, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/73b56de0b6353db0bd5025fd60a64bea6a7beb61b85459eb7aee8fe261ef2815.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": "74193a791e87b93afc08b526fd5b56192a8809549c7d16c80cd675ca005c0157", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae5b8e1a4b93f9e6b9aab2440cacfaeed55cbca3dc30e5c38bc38c0dc6a60a1e", "generation_seconds_batch": 16.07405153394211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.0146708488464355, "genuineness": 4.042920112609863, "identity": 0.49230214953422546, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/74193a791e87b93afc08b526fd5b56192a8809549c7d16c80cd675ca005c0157.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "7456274a5e38ca2c1075334a8c651f984240f57893779e8c3ba68e896230f470", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2887f19f9ea08006617ad91a1ded1ac852a3218a307785f82a242725c306da06", "generation_seconds_batch": 21.2765400740318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.8421617746353149, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7456274a5e38ca2c1075334a8c651f984240f57893779e8c3ba68e896230f470.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "74a5d846700373bbdca492f185134961f42d2197cc57821862060296bceba7bf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db492750f2b2bd33cb0e35f72188ef7c71c3138590e32dfb88993aac0bdab9f9", "generation_seconds_batch": 20.292147210100666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.173524379730225, "genuineness": 4.18726921081543, "identity": 0.5459492206573486, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/74a5d846700373bbdca492f185134961f42d2197cc57821862060296bceba7bf.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "74f805fc4806707b2f5c952485353ea7d6b0a9c476b2764db9a424a03b19e35e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "81b29e202917a71ff626aef3b49f34cf52d0fab0cab2a0465b698fdd86e9d6ce", "generation_seconds_batch": 25.673564390977845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 5.863759994506836, "genuineness": 4.093128204345703, "identity": 0.6881536245346069, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/74f805fc4806707b2f5c952485353ea7d6b0a9c476b2764db9a424a03b19e35e.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "7506705746f7ef76ff3c2d38807dfc3141407a61096166d7818c7627a12498e1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "47ae0b48640c303f1da6063a0e6bc106add2fca5dc62dc1c7c315c52b5c4a5d2", "generation_seconds_batch": 20.813051688019186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.4687501403907581, "ce": 6.149696350097656, "genuineness": 4.482243061065674, "identity": 0.5624717473983765, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7506705746f7ef76ff3c2d38807dfc3141407a61096166d7818c7627a12498e1.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "75087bed48247ac446ba85658415310e63c28c2bfb528ad44e5b7d76d971391f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "509bbea16bbc13e846d6961bde3726cf494f8c6eae00e965147c68cbefdf5399", "generation_seconds_batch": 22.34909993398469, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.445312509236234, "ce": 6.173524379730225, "genuineness": 4.469690799713135, "identity": 0.7876782417297363, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/75087bed48247ac446ba85658415310e63c28c2bfb528ad44e5b7d76d971391f.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "75dd5ea9a367555906b29e1404b2733151b8de064b5f44fd3fa2259722461730", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "76168d6108c4ae7e994c51419cd5029659dd4b941e3b77d2c633f85321280bb9", "generation_seconds_batch": 20.323362982016988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 1.1015626348490175, "ce": 6.300607204437256, "genuineness": 4.3127899169921875, "identity": 0.5364668369293213, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/75dd5ea9a367555906b29e1404b2733151b8de064b5f44fd3fa2259722461730.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": "76189234b06da57b08cdfafd8163b80327362b2f2d1b95e26afd41f5c0bfc446", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60dabfeff2e5e985dcf21d4f7994f4deed9448d595f49aa60e9d145706ea6ef7", "generation_seconds_batch": 14.404132487019524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.09409761428833, "genuineness": 4.15588903427124, "identity": 0.7310081720352173, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/76189234b06da57b08cdfafd8163b80327362b2f2d1b95e26afd41f5c0bfc446.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "76422fd8b07e0fc3716bbdce44e3c7cab3f7a578e82217ed9315db4acd248694", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a08ccf31ad768c9c49b7d8b7c06a1f863a31109c5adb361e972bdd738afdf42f", "generation_seconds_batch": 21.820332500035875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.348263263702393, "genuineness": 4.105680465698242, "identity": 0.3921172618865967, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/76422fd8b07e0fc3716bbdce44e3c7cab3f7a578e82217ed9315db4acd248694.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "76f293d9b9d572633ab89b41472c5f14f6db15befc8cf3065a4400f83828a090", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "862f07a817aee7333d5ed9108873f9d468df01060d588e462a1e6100b310cb89", "generation_seconds_batch": 21.07782974501606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5531973838806152, "target_z": 1.2734373355950335, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.6777547597885132, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/76f293d9b9d572633ab89b41472c5f14f6db15befc8cf3065a4400f83828a090.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "777b79f66460cf81b78a831c82e42f8081319066083f581f2411eb1702bf68f6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "58f3dd7db69922efe17be515a12e393355540ea5e31b5e3a3bf18abf01829c81", "generation_seconds_batch": 22.433946365956217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.006728172302246, "genuineness": 4.281409740447998, "identity": 0.7230530381202698, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/777b79f66460cf81b78a831c82e42f8081319066083f581f2411eb1702bf68f6.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99eb663ac0f7ab94dfeafa2ae21ab259afbb1fa7acbc33d0d3afec6f5ee16a48", "generation_seconds_batch": 16.56279383704532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.709154486656189, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "786b70864672934bc8c5c3e840fe0dfe634bf706f12310e4daaee3dcb6f84c88", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a78d267b9ce801d46cc9a3fe6d2c623bfe7c6863176837853817a9bb9d9826bf", "generation_seconds_batch": 24.283673704019748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 6.38003396987915, "genuineness": 4.20609712600708, "identity": 0.6004928350448608, "hypothesis": "Ik heb hem het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/786b70864672934bc8c5c3e840fe0dfe634bf706f12310e4daaee3dcb6f84c88.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "787ac92688c7685101621ac25e68aeadca24ee4902e57fd8f1624ce9cee6cd90", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "688d9f410a7f75f3d3124507058cdc0ea4cde4251768e632b5f77785edd7a5de", "generation_seconds_batch": 22.04793039895594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.221180438995361, "genuineness": 4.143336772918701, "identity": 0.5515912771224976, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/787ac92688c7685101621ac25e68aeadca24ee4902e57fd8f1624ce9cee6cd90.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "78da688de49bc5b490a3ffdeaf74ae8e972a974e7b4b5a015f2c0c0a190a026f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ee45353321bbf4ffe146c340e16e62960c3fd4b91d5b8c846449e3cd6e0d83cb", "generation_seconds_batch": 21.030747838085517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6796874020959187, "ce": 5.95112943649292, "genuineness": 4.268857479095459, "identity": 0.7454037070274353, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/78da688de49bc5b490a3ffdeaf74ae8e972a974e7b4b5a015f2c0c0a190a026f.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "794cbd35e08584ac210e92358ef717d70fc7e13f8d857dcd4fb843112dd950bc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "48aa7d17c5119d7407899e2431cdfdf75e58c7e260a1c639991c6ce95734e69d", "generation_seconds_batch": 26.68020153895486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1834073066711426, "target_z": 1.8984373651509825, "ce": 6.316492557525635, "genuineness": 4.268857479095459, "identity": 0.512807309627533, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/794cbd35e08584ac210e92358ef717d70fc7e13f8d857dcd4fb843112dd950bc.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02cad72e1c42fe847b483fce6cb35cf914ca96b22b308be3845b19836508635d", "generation_seconds_batch": 15.6802391349338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.9609375572646512, "ce": 6.09409761428833, "genuineness": 4.519899368286133, "identity": 0.7182250022888184, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "7971a2897a964e3c614287e244e6fec1090c0e1a7c641e9f2d88b9cee814987e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4300f855c39d893655a0e060faf8b8791e4e855dbabee9457dfd03b3516a04d7", "generation_seconds_batch": 21.607595283072442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.157639026641846, "genuineness": 4.519899368286133, "identity": 0.7188965082168579, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7971a2897a964e3c614287e244e6fec1090c0e1a7c641e9f2d88b9cee814987e.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": "79b071cb5900d6e8b315e0f7c054c90066b2e1123df1b33a73ddda77931d6c75", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67845b2770be9fe4c5495479d0c1a19c2570e0652d3d9fb7d7d7ba6396ed9fef", "generation_seconds_batch": 17.286179161979817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.221180438995361, "genuineness": 4.356721878051758, "identity": 0.8107070922851562, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/79b071cb5900d6e8b315e0f7c054c90066b2e1123df1b33a73ddda77931d6c75.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "7a0e35216a3c0c9a8961c7bcf04693438202ba6e0985bc1b0036179f3fa2801d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "65a2b7257789e5ff9adc31de1ffbf785b2cf2f12dc9cd70e1314da979f7f6835", "generation_seconds_batch": 22.158739043981768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.332377910614014, "genuineness": 4.017816066741943, "identity": 0.3588508367538452, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7a0e35216a3c0c9a8961c7bcf04693438202ba6e0985bc1b0036179f3fa2801d.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": "7a600c47a8a73f759fc55c683cbb0cf30918aa76d1c9973e6943e17218e8e44e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42bb72fefe88e2d17c224eb9ccae87ee53ae0bb8d18e22cd8b63d4eb3c1a652c", "generation_seconds_batch": 21.9024108099984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.5385167598724365, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7a600c47a8a73f759fc55c683cbb0cf30918aa76d1c9973e6943e17218e8e44e.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": "7a7c7db10b5ce3b7bc6b5a69be3b4889715c39aca8ae62aface553f89a58d98c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "161087c4becf5c4bd9ced3f93f2c33c1929b69cc7d7b7812ea3259d82be345c6", "generation_seconds_batch": 16.920426217955537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.2187498448312675, "ce": 6.0305562019348145, "genuineness": 4.482243061065674, "identity": 0.6094595193862915, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7a7c7db10b5ce3b7bc6b5a69be3b4889715c39aca8ae62aface553f89a58d98c.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "7af76df5645ab35b288b1d26895133332c4869bea57cd222959f8cce0e85c695", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "25784f54b7721037783bbf756aba9292f72e1c4e4d8ef6d5584300a890eb91ab", "generation_seconds_batch": 21.5460973510053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.109982967376709, "genuineness": 4.595211505889893, "identity": 0.6522125601768494, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7af76df5645ab35b288b1d26895133332c4869bea57cd222959f8cce0e85c695.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "7bb3c40e773c0370e1c408139c465438fbaedcf7929d27ed51bad9b4ec0aa798", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "8a7b10843aec605420737e2290bb4988ac685c921cfb7a60475dc900b8b4b92c", "generation_seconds_batch": 19.420162008027546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.675555944442749, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7bb3c40e773c0370e1c408139c465438fbaedcf7929d27ed51bad9b4ec0aa798.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "7bcf1e7a233cd5369e01002c42fd6024d452488ef03b4f1528891758321ef1f9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b1cd7424b6111420159bbb6263d4bfdd043cc099b32588cff7340fc890940abf", "generation_seconds_batch": 15.240477417013608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.6754530668258667, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7bcf1e7a233cd5369e01002c42fd6024d452488ef03b4f1528891758321ef1f9.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "7ca13ae0bf6ae0c11a89935485d86528b490929bd2f06645d1d22c23352de352", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "23ecaa2b495f735c4899936c885f26b4f50d8c3745a2832378321c4a391603cf", "generation_seconds_batch": 22.333251575939357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.1894097328186035, "genuineness": 4.457139015197754, "identity": 0.569800078868866, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7ca13ae0bf6ae0c11a89935485d86528b490929bd2f06645d1d22c23352de352.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "7cd28a85d97d1c0330fe6dce2e942aa35d8c697472e374e5d2fc070cba7907ac", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "04032fd13e91618d604c14466f95bf7caf9fd9f1a111be27c20127ec1e04656d", "generation_seconds_batch": 16.329043647972867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 6.403861999511719, "genuineness": 4.2374773025512695, "identity": 0.39425256848335266, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7cd28a85d97d1c0330fe6dce2e942aa35d8c697472e374e5d2fc070cba7907ac.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "7d517e02150bc2eac4f9c289161fbda4ff2e47436c0f3eb9c1f5875426cf0f0c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0a6beecc2cfd16ea3f3ec4a85a02ac783966226b3aae9b33f2258421a0e4b15", "generation_seconds_batch": 20.72027807997074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.293236017227173, "target_z": 1.0156250480284172, "ce": 5.982900142669678, "genuineness": 4.2249250411987305, "identity": 0.781953752040863, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7d517e02150bc2eac4f9c289161fbda4ff2e47436c0f3eb9c1f5875426cf0f0c.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "7d62f94c7f50c099c5f12d721dbe4814f77bd658818ab5aea3b0b1746ecd0383", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "64f37d26fb96915600474054e911797bf50d5bab870ae82181c18015dcc0a2ba", "generation_seconds_batch": 13.660180047038011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.221180438995361, "genuineness": 4.325342178344727, "identity": 0.637033998966217, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7d62f94c7f50c099c5f12d721dbe4814f77bd658818ab5aea3b0b1746ecd0383.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": "7d9b9e3060c01fa3ac6803b5bf4bf314a2f66931246091e559990f06440c99b7", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ee3cb4aa5fd737adc02d4079724f2bf6e3b6e82bb2a17a20523437554c94d5c", "generation_seconds_batch": 15.987199209979735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.483288764953613, "genuineness": 4.118232727050781, "identity": 0.013776704668998718, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7d9b9e3060c01fa3ac6803b5bf4bf314a2f66931246091e559990f06440c99b7.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "7de8a56631771731df8cdedb8865428e33fa1d1723ea6eab68c0a30fd74337c6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efacdf52dd5d2ecefc32ab76b7434abb7d469b57cc16c47f55ace2ed0044f65b", "generation_seconds_batch": 21.163426148006693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.078212261199951, "genuineness": 4.256305694580078, "identity": 0.7043628692626953, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7de8a56631771731df8cdedb8865428e33fa1d1723ea6eab68c0a30fd74337c6.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "7e3fd5fcf1f3919881862e6a135310de5773ef9e10943eb46763d2dac3a763ca", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "14d78d488428406641829243433860a89ccd1a9b48c661baf04d441bc0cc3f21", "generation_seconds_batch": 21.53510856302455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.086154937744141, "genuineness": 4.306513786315918, "identity": 0.628183126449585, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7e3fd5fcf1f3919881862e6a135310de5773ef9e10943eb46763d2dac3a763ca.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "7e7b8cb3615a2f7297c1b019a9dcc8b42591e8aa956c029047e6ca2f03364b44", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "c8e2595348daffb1fd3ea33a5fe26d7281be6089e4ff5e03e30ef1ce3afb09a8", "generation_seconds_batch": 20.305099076009355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.5168991088867188, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7e7b8cb3615a2f7297c1b019a9dcc8b42591e8aa956c029047e6ca2f03364b44.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "7f446fe64b3fb025342c82bf7c9df9dbdb1b7f440a6d21663160e86bc0e569dd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8e6310a635820bf4c73f548ce37099d821f64ed80378f4c4d04d19df9ffb4e5", "generation_seconds_batch": 20.044202320976183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 5.95112943649292, "genuineness": 4.3127899169921875, "identity": 0.8198419213294983, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7f446fe64b3fb025342c82bf7c9df9dbdb1b7f440a6d21663160e86bc0e569dd.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "7f5cd99aadd5e10f73508a25909c780c17498bdba4e9e56183a20cd1ed80737c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0ad67f3087f12cef346ab827f5ff4c5bccab1e2f7496bb2da7a79eafa9b370c5", "generation_seconds_batch": 15.240477417013608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.0146708488464355, "genuineness": 4.218649387359619, "identity": 0.7235944271087646, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7f5cd99aadd5e10f73508a25909c780c17498bdba4e9e56183a20cd1ed80737c.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff71071095f2befa2aeb92bd2ff29b5faa119db4eff23cbe5599c782f9d8450f", "generation_seconds_batch": 16.10312503296882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5847082138061523, "target_z": 1.3046876680994604, "ce": 6.403861999511719, "genuineness": 4.3504462242126465, "identity": 0.4289855360984802, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "7f99cd41041886011b4ec15620475010812ea78cac060a2f6cf3cde651d993d9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "adaf412b6ecb462e7167692a0b1c02df8da6fa426f71114feb41b61679e6ae96", "generation_seconds_batch": 16.91973064094782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.986008644104004, "target_z": 0.7109374981527532, "ce": 6.4515180587768555, "genuineness": 4.595211505889893, "identity": 0.2097887098789215, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/7f99cd41041886011b4ec15620475010812ea78cac060a2f6cf3cde651d993d9.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": "8005495e8748470cd71f6c865179c2b180c016e6bdb909a843897d8ea6c96830", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3d717fff1ba42592cbafd048ca5c00287ab4cd1b3b79b450ad97a70225c280f", "generation_seconds_batch": 15.845448528998531, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.403861999511719, "genuineness": 4.193545341491699, "identity": 0.3130364418029785, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8005495e8748470cd71f6c865179c2b180c016e6bdb909a843897d8ea6c96830.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "8062f34b12c3169ffe8ed09ce10dac7ff53bf05906575c5ccd3d34172556b784", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "68a5018fa5d15dd9972f446df2978883022b8c5ecb3c244989f2ba91caa6debd", "generation_seconds_batch": 20.292147210100666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.125868320465088, "genuineness": 4.3504462242126465, "identity": 0.6381901502609253, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8062f34b12c3169ffe8ed09ce10dac7ff53bf05906575c5ccd3d34172556b784.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "807147208031cf81f782a5198054b8db73d87c06191fa3fb85ca47fd0197665c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8a423828bf94022fceecd05a6f8b95554f495745b3cbe1d0d549717e62f8fc93", "generation_seconds_batch": 22.606556402053684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.3984374833747786, "ce": 6.38003396987915, "genuineness": 4.413206577301025, "identity": 0.3512430191040039, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/807147208031cf81f782a5198054b8db73d87c06191fa3fb85ca47fd0197665c.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "807aff71afdec1f3dc0ab8ecda8c575ed947b6b32b545057bc1a66192bcc21cd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5bde3fbb904027b9a7b5e04d58858ffdb935f57eb4b00c758446a86ca29d822f", "generation_seconds_batch": 22.40669529500883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9623757600784302, "target_z": 0.6874999852220255, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.6243853569030762, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/807aff71afdec1f3dc0ab8ecda8c575ed947b6b32b545057bc1a66192bcc21cd.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "80bad8cd84eb8276e4682b92cbd0a55947da36dfda68d77484ad0a7da72889a0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9ecd6403430f055980847df0b3fdf454259442b247f4e49f6682f28848760c93", "generation_seconds_batch": 22.40669529500883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.0146708488464355, "genuineness": 4.532451152801514, "identity": 0.6531490683555603, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/80bad8cd84eb8276e4682b92cbd0a55947da36dfda68d77484ad0a7da72889a0.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a691eb8612da1e056b48245c005b1638fdf6a076f2441d1feb4351454a79137d", "generation_seconds_batch": 15.219675200991333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1514387130737305, "target_z": 0.874999970444051, "ce": 6.24500846862793, "genuineness": 4.344170093536377, "identity": 0.658647894859314, "hypothesis": "Las llaves están en la mesa de la cocina. Mire en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.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": "80e4e211384daa18cb0edd4c19ab9f06d0203554d71ddf1b5c8cf20cc80b920a", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "330c4d3c3a37b1ce136d8f2d2ea0fe24cc850555211ce505b192978911be649b", "generation_seconds_batch": 17.073276616982184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1199283599853516, "target_z": 0.843750110834809, "ce": 6.24500846862793, "genuineness": 4.218649387359619, "identity": 0.5531930327415466, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/80e4e211384daa18cb0edd4c19ab9f06d0203554d71ddf1b5c8cf20cc80b920a.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "81692249f1c3e37e2c9faba6ce75eb5a8a004ffba241204692ea5662704e6e15", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ffa9334abf17c36a3d7a2e8d36b1af027499f441940f6d69254ace1af0bac222", "generation_seconds_batch": 21.996487003052607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.624999911332153, "ce": 6.229123115539551, "genuineness": 3.9801597595214844, "identity": 0.29100027680397034, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/81692249f1c3e37e2c9faba6ce75eb5a8a004ffba241204692ea5662704e6e15.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "82161ab4ee6dea30f5cc578849658b12531a4197a39797e154edae56a6ebdc89", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d3bbda018e45e37946947bb958b08960b67af2c24eb77b4f756927315486914e", "generation_seconds_batch": 22.002071999013424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.324435234069824, "genuineness": 4.319066047668457, "identity": 0.15453502535820007, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/82161ab4ee6dea30f5cc578849658b12531a4197a39797e154edae56a6ebdc89.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "821b8b8cf13b6b2f6acc7a7861c3aae599725ad4ab72b37cf0d1d5eadd8c8ed4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "206d9a57b1190fff07e44f3316aafe9f2e7bcf0103890a46cfa7165d1a07978a", "generation_seconds_batch": 20.832028696895577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.173524379730225, "genuineness": 4.21237325668335, "identity": 0.4702555537223816, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/821b8b8cf13b6b2f6acc7a7861c3aae599725ad4ab72b37cf0d1d5eadd8c8ed4.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "8270fe0242213b3111ce4aff34c41a0c61268b887e45357bda4584592f327818", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b6d0f0440dae4b8badb57871ba162d7149f9b5ef2bc8a1dd66fe6dd38aa82d7f", "generation_seconds_batch": 27.232282888959162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.7943031787872314, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8270fe0242213b3111ce4aff34c41a0c61268b887e45357bda4584592f327818.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "830f765d8301c134707c2f00dad63f4413233852ff6d441b26d5e327812740cc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "5ae590e3ef76bfe13ffe836a72939fe935e974c90f8eed32af95855a9415ffa3", "generation_seconds_batch": 15.391588304075412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.0146708488464355, "genuineness": 4.319066047668457, "identity": 0.5003386735916138, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/830f765d8301c134707c2f00dad63f4413233852ff6d441b26d5e327812740cc.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "83d231bd5fc9b51f90a98fb0a60fdee8d1f429de302cea816af0bd73e4ac74c5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b3809b810e4a46c3d3e057217b39b31383d8633b04d686bfcf6742b34678ba18", "generation_seconds_batch": 22.86087413295172, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.229123115539551, "genuineness": 4.124508857727051, "identity": 0.18390260636806488, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/83d231bd5fc9b51f90a98fb0a60fdee8d1f429de302cea816af0bd73e4ac74c5.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "83dad72e206cf77c3301343062258bdaaa1c8f658c8dcd660c56009a824a6518", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "de12e4afa1a6ca5639edf28e35807ae07f8f8c8d703ae06aa6e419f00df9ec5a", "generation_seconds_batch": 22.618444842053577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.117925643920898, "genuineness": 4.15588903427124, "identity": 0.6883481740951538, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hace.", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/83dad72e206cf77c3301343062258bdaaa1c8f658c8dcd660c56009a824a6518.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "83f9f8c44624b044fd57fea660f2fcd64003675ade26257a779344a864d6e65f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9c714983c98ac454f6162529587b7af196095069daed0626e8fab9d386742ec4", "generation_seconds_batch": 20.323362982016988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.268836498260498, "genuineness": 4.218649387359619, "identity": 0.5638397932052612, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/83f9f8c44624b044fd57fea660f2fcd64003675ade26257a779344a864d6e65f.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "83fd3f33873c0a47694e6df2cb64002477089bb77d09d93175d420f8b4df5ab3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "490452b2dc5d67b3a1de9d17de4cd00a865e6242ec0d6162c139d7304731e5b6", "generation_seconds_batch": 15.442915622028522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.139622211456299, "target_z": 0.863281154866789, "ce": 6.348263263702393, "genuineness": 4.532451152801514, "identity": 0.2726198434829712, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/83fd3f33873c0a47694e6df2cb64002477089bb77d09d93175d420f8b4df5ab3.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": "84163adee87fe877d70ec783dc33b868e9927220148629968aa0ade52e50e24f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cf0a74acdb14f69bb49c630ce71c3a550c0cf213a8cb9ac62c1aab323bbba32", "generation_seconds_batch": 16.54500030190684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.046441555023193, "genuineness": 4.369274139404297, "identity": 0.7769813537597656, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/84163adee87fe877d70ec783dc33b868e9927220148629968aa0ade52e50e24f.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": "8448e777e2f0f3ea40c409c2c2fc07b47816b9358a9e7df59331b05783266ade", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e5f4a52c094d24aec4b02bb2d02e6794236f7912d44d6c1f94a756606a45141", "generation_seconds_batch": 22.51586488704197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.445312509236234, "ce": 6.443574905395508, "genuineness": 4.231201171875, "identity": 0.3531281352043152, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8448e777e2f0f3ea40c409c2c2fc07b47816b9358a9e7df59331b05783266ade.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "84581c6c8dca0107dd3b0a21b70cf988ce03c293c6a7afdebedebf7a6409d35b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "2853f6e5024f42bf90348e03e1c63f2dd84f195a029467824ed89c15fd3d945d", "generation_seconds_batch": 22.466701232013293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.046441555023193, "genuineness": 4.306513786315918, "identity": 0.48341095447540283, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/84581c6c8dca0107dd3b0a21b70cf988ce03c293c6a7afdebedebf7a6409d35b.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "848fad9eba62651835cc63a97b0750515ed440f0bf5fbe9b3011a437f6379ce0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b759513379c91f62234b392c9b151b8835ce0b2f6bb449605383dd17493267b6", "generation_seconds_batch": 16.34311248292215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.133810997009277, "genuineness": 4.174716949462891, "identity": 0.7802625894546509, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/848fad9eba62651835cc63a97b0750515ed440f0bf5fbe9b3011a437f6379ce0.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "84b0b42cd0c2091385a500b02bd357c1da6a266155ad6876e9f77d9a238c3cd1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 130, "seconds_generated": 10.4, "mp3_sha256": "c1455650b670c462c552867a55085cfcf87942e44b29e84e264f758ede9ac976", "generation_seconds_batch": 27.433922859025188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.234375011083481, "ce": 6.086154937744141, "genuineness": 4.030367851257324, "identity": 0.7146044373512268, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/84b0b42cd0c2091385a500b02bd357c1da6a266155ad6876e9f77d9a238c3cd1.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "84e33f843bbed2038cf7708065da26830e915820fd66a1c96b128bedaf4b8740", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4466db9bd1e05437ea000d2bcc5bb56a78680d6f34b3dcb24937acb3905b3f34", "generation_seconds_batch": 23.16196779801976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.203125151474239, "ce": 6.38797664642334, "genuineness": 4.256305694580078, "identity": 0.587931752204895, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/84e33f843bbed2038cf7708065da26830e915820fd66a1c96b128bedaf4b8740.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "85662f34a8e4ae1e36245d9ad15cd13ee11c2a876710a158c8d69655422a9aab", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9ff570c91ca0bfdfed360b14c97ef35d74d1d5717bdfe9cbe1d657097ac662c3", "generation_seconds_batch": 18.13981138705276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3956451416015625, "target_z": 1.1171875646536387, "ce": 6.316492557525635, "genuineness": 4.080576419830322, "identity": 0.40205711126327515, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/85662f34a8e4ae1e36245d9ad15cd13ee11c2a876710a158c8d69655422a9aab.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": "859c4afb1f4738620c66120b547960c977d3c53f2feab661cc0fe4d3f7f3818c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fbb043c11804f2d772eaaf5f8e66a0abcdb9ddbebad9d1d1122bb997932ed70", "generation_seconds_batch": 15.265560391941108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.6838918924331665, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/859c4afb1f4738620c66120b547960c977d3c53f2feab661cc0fe4d3f7f3818c.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "865a92e3ac54feb4f2bcf1a1274fbd3849e0893386943b7e84b5ebe001bcaa4f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "79b1594a3b00384591d75db8938f449fc653d49910616dd90a6f241220fd3199", "generation_seconds_batch": 22.002071999013424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.100234270095825, "target_z": 0.8242188303552365, "ce": 6.284721851348877, "genuineness": 4.143336772918701, "identity": 0.32847267389297485, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/865a92e3ac54feb4f2bcf1a1274fbd3849e0893386943b7e84b5ebe001bcaa4f.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": "86c123416236edec617766428a03fa84a3a526fdb5a9008331053a2636c8f5af", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aba824c2e3ddb2b8be1c8febad12d33cdced34b5a364218eb9fb6a491d4ca6fd", "generation_seconds_batch": 16.0294995669974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.419747352600098, "genuineness": 3.9550557136535645, "identity": 0.3090570569038391, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/86c123416236edec617766428a03fa84a3a526fdb5a9008331053a2636c8f5af.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "86caa0a93d46e6eb34caf2e3758129aaed2a7c86a7bf4116eaeb87ef33329fc1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f6f17bf861c7429dc4b2d55c2fe7ca79bbb69ccf3a8e332eb5b103e1acb4f489", "generation_seconds_batch": 21.98617203196045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9374999261101273, "ce": 5.974957466125488, "genuineness": 3.8671910762786865, "identity": 0.7551110982894897, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/86caa0a93d46e6eb34caf2e3758129aaed2a7c86a7bf4116eaeb87ef33329fc1.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "872d54d29f68b8f5cfe2373588ecb62bd1337d452fade98176e3856c2aa89e1a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2f489c5a236a99484e589d82254610b75163cd15dd090276a481dddb6b6ad7a8", "generation_seconds_batch": 22.849052945966832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6796874020959187, "ce": 6.173524379730225, "genuineness": 4.149612903594971, "identity": 0.5833313465118408, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/872d54d29f68b8f5cfe2373588ecb62bd1337d452fade98176e3856c2aa89e1a.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "873297f54ce04cf721c1b077c21fd6d1c612bab7ce6d1874a2e84dbf5c85b701", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "bf7d8477afd802d4f16c684fe8aaadf3391ddf9e7c6e4edc06d9e892bca7a167", "generation_seconds_batch": 21.411273667006753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0844788551330566, "target_z": 0.808593664103023, "ce": 6.181467056274414, "genuineness": 4.306513786315918, "identity": 0.7545185685157776, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/873297f54ce04cf721c1b077c21fd6d1c612bab7ce6d1874a2e84dbf5c85b701.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "874cab117b2c7d595d1de779ec2d8b9c5e5c0a27f3fb33d37df883ab86ac8314", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e3eeffa6661966fbc730c14c328ef257b772267256fada87aeb9a7320ba1fba1", "generation_seconds_batch": 22.76982721593231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.9843749519715828, "ce": 6.4515180587768555, "genuineness": 4.344170093536377, "identity": 0.6051799058914185, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/874cab117b2c7d595d1de779ec2d8b9c5e5c0a27f3fb33d37df883ab86ac8314.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "8798308b477003620d939136bee79fda363d7cc43ffe1090a5d42a92ac98805c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d4472bd75e3723157b9a105e67109b4b0f3f1ec10f04c3a479d31a86d1ae8136", "generation_seconds_batch": 21.916390803991817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.0305562019348145, "genuineness": 4.570107460021973, "identity": 0.6489419937133789, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8798308b477003620d939136bee79fda363d7cc43ffe1090a5d42a92ac98805c.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "87b55c4f4298fb0eb32467f5a2e33a01eb6874982e05adba23561bd2ccc0e3cd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f53766877c00090524da5f36b09974f7b6ad4c26abc434a816d2e481c534c410", "generation_seconds_batch": 21.90557630197145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.156249889165191, "ce": 6.348263263702393, "genuineness": 4.231201171875, "identity": 0.535077691078186, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/87b55c4f4298fb0eb32467f5a2e33a01eb6874982e05adba23561bd2ccc0e3cd.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": "87c04fe0e2fadfed95d3c12b176d3a346a55564d9e03cc1f4e6eaa6da09cb1fa", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8e4f21f82eb451676ee3c5c863ccf9eea8107d6e3a1d3c11d257c89a71eba64", "generation_seconds_batch": 20.71616707299836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.427689552307129, "genuineness": 3.898571252822876, "identity": 0.3701406717300415, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/87c04fe0e2fadfed95d3c12b176d3a346a55564d9e03cc1f4e6eaa6da09cb1fa.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "87d7d8b4df39d6a3bada637bcb2c30e5a3174d44a57e8f8a5ffd3c5a54168973", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "61b9d474e71d25fa78c976b942cf7d66bc4a51072bf7f17da63004392cba487c", "generation_seconds_batch": 22.265967768034898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69545316696167, "target_z": 2.4062501847246818, "ce": 6.300607204437256, "genuineness": 4.093128204345703, "identity": 0.5168328881263733, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/87d7d8b4df39d6a3bada637bcb2c30e5a3174d44a57e8f8a5ffd3c5a54168973.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "881d187d2957c2835afad8f8a65c7ff490b2e654d445eb5bdf3ef05940d40639", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "14e12df51826405a811ee38ea9f8ec27054741228f141a86ede7dbb60c99f9db", "generation_seconds_batch": 16.549790724995546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.173524379730225, "genuineness": 4.494794845581055, "identity": 0.7178783416748047, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/881d187d2957c2835afad8f8a65c7ff490b2e654d445eb5bdf3ef05940d40639.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "8823d4b5a3ff06a58b1ac68e9ecc1ff777b11c228c1f3bd2b31fbcca74d051a5", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "35ed9aaad7d89311a48b7c04e0ec3c588e66758d8433e605f613bd454e60706c", "generation_seconds_batch": 18.20241677504964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.348379373550415, "target_z": 1.0703125387921832, "ce": 6.300607204437256, "genuineness": 4.105680465698242, "identity": 0.41465142369270325, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8823d4b5a3ff06a58b1ac68e9ecc1ff777b11c228c1f3bd2b31fbcca74d051a5.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "882ef2b81b2c2661a5933e7a3f11537b5b3ef8f1d792e29ff09430abe5596c7f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bec63d225f499620ca97acb2b45ca5581fa6477cfd48ff01ce78fcc2552a3312", "generation_seconds_batch": 21.183293274021707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 5.967014789581299, "genuineness": 4.218649387359619, "identity": 0.7027358412742615, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/882ef2b81b2c2661a5933e7a3f11537b5b3ef8f1d792e29ff09430abe5596c7f.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "88991347d520836901841a5335f32bc22df3b3e951889f7b9670e289d4d854d1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b68b9b76b8aa265e57b1076ec76a4ca42a2dba81f4578ec7d68ba4c6812cdc7d", "generation_seconds_batch": 22.673301325994544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.922987699508667, "target_z": 0.6484375424866767, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.7294962406158447, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/88991347d520836901841a5335f32bc22df3b3e951889f7b9670e289d4d854d1.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": "88f0c6e396b1cdc27e04495049880ce464fd59a7dd6d5746e016f57c4d52b0cb", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I wrote the number on the envelope.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "744a8628f3d406b81be4addf53e1de563a406c9172c7afb674993725e4073431", "generation_seconds_batch": 21.0285669779405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.507116317749023, "genuineness": 4.068024158477783, "identity": 0.2895732522010803, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/88f0c6e396b1cdc27e04495049880ce464fd59a7dd6d5746e016f57c4d52b0cb.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": "896362ccd4cf88b34b0730f4c8ea798e48457de570d1624cc0a22c3c80ea941b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "500b2c887ee3aaac643b13bf15955ceaf9640608987e53fb45812cacfdf050f9", "generation_seconds_batch": 15.757981001981534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.395919322967529, "genuineness": 4.557555198669434, "identity": 0.5502709150314331, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/896362ccd4cf88b34b0730f4c8ea798e48457de570d1624cc0a22c3c80ea941b.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "89996abef6084e897da5ff7da622c2e159f8c77dd477d978d70cca0c191a7602", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6041501599726e72a9c5e7d0931b442fa2f700fb313c700652f0ef34da12cc1d", "generation_seconds_batch": 16.564654507907107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.157639026641846, "genuineness": 4.394378662109375, "identity": 0.6049627065658569, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/89996abef6084e897da5ff7da622c2e159f8c77dd477d978d70cca0c191a7602.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "8a2378a8cbb5008faee36e16c6fa6f7ef49a019494c6398588c1613aa78b8d46", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f663f57b58be1171e7f1c28f6f6af7d5da5c1de8386c3f6da6cddd6e2c9fea87", "generation_seconds_batch": 20.231525473995134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 5.911416053771973, "genuineness": 4.080576419830322, "identity": 0.6145597696304321, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8a2378a8cbb5008faee36e16c6fa6f7ef49a019494c6398588c1613aa78b8d46.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8a26a6d530ec002cf517fe4331a2de7cb96dfed1de7bae55044db13f19d450ee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "215dec1b200486e8fdd220bc66dc7789054ff24b5ea615bb36501b0e18406021", "generation_seconds_batch": 16.736632735002786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.41180419921875, "genuineness": 4.557555198669434, "identity": 0.33968397974967957, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8a26a6d530ec002cf517fe4331a2de7cb96dfed1de7bae55044db13f19d450ee.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "8a3c7345c46421bd4a4ff259c26b7a468ba2726d056fbdd717f17b67c61c6162", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "85ce59735c899cf679aaf5d647ea1e3f8b0d928bdaaebffb44ad66fbc85000c6", "generation_seconds_batch": 24.07217539602425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.7782199382781982, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8a3c7345c46421bd4a4ff259c26b7a468ba2726d056fbdd717f17b67c61c6162.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "8a8ca18a69a4066916065d8a54bda17bddbc8f7577e0743a227222b3fdc742bf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "86b9f03b7ac28bb087dfafef7ecc0757e85f3da88a97073e9b05f87af0ecb347", "generation_seconds_batch": 22.282496836967766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.6687651872634888, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8a8ca18a69a4066916065d8a54bda17bddbc8f7577e0743a227222b3fdc742bf.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": "8ac97fd3d7564e618d8a9a6714d7928740afd495bcef8f6f42f252dbef8f6b53", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c776d64684c8bb87489b5dbc0b293dd4defa958ba54224b339ce75c3b89ba3a", "generation_seconds_batch": 15.464171582018025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4591240882873535, "target_z": 2.171874818969812, "ce": 6.284721851348877, "genuineness": 4.268857479095459, "identity": 0.67741459608078, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8ac97fd3d7564e618d8a9a6714d7928740afd495bcef8f6f42f252dbef8f6b53.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "8b4d58a348c21081b1cbc4f1d3fa081e81b4c63901ec81af1403b198d96df739", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "50930ecdf09b466535cf0e06ffa635914513861b8e162ba746082599f16604f9", "generation_seconds_batch": 21.664702587993816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.849419891834259, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8b4d58a348c21081b1cbc4f1d3fa081e81b4c63901ec81af1403b198d96df739.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8b6f8bde98ab08f52271ca1ca4017196881f4824968239bf5a21b0c38f7ab40a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "63ee4384fb8de89c519c69fb09856519268acc6a77e50336012ca893ddcad095", "generation_seconds_batch": 20.299461154965684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5940968990325928, "target_z": 0.3222656550177608, "ce": 6.340320587158203, "genuineness": 4.532451152801514, "identity": 0.40302664041519165, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8b6f8bde98ab08f52271ca1ca4017196881f4824968239bf5a21b0c38f7ab40a.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8b8baf7c168ae261a14b7a1dcd14f273d48f272cd6c052e5424076fe563e60b7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "57a14aa29340213d0ae79a661b9f6798628856e8b9c78c81c725005ebb47508a", "generation_seconds_batch": 22.98339196201414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.36217477917671204, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8b8baf7c168ae261a14b7a1dcd14f273d48f272cd6c052e5424076fe563e60b7.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": "8c8867904b5ae59e6db92aa20029b78cf8b1593c4efbb2a3555425c3c786ca3d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "827668d6ea04e9411b9c2ac0d8b09c59ce7627ff8d3e146600b8f82badddebca", "generation_seconds_batch": 22.190684631001204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.372091293334961, "genuineness": 4.331617832183838, "identity": 0.6322653293609619, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8c8867904b5ae59e6db92aa20029b78cf8b1593c4efbb2a3555425c3c786ca3d.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": "8d551a5d6572a961d7db0946181829576e3549da645822ccbe5f7335b47f86bb", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "763564a3064efa4c37d5cd35c3f0610c1dfd9c2ecca12c8edc0a348c74491c15", "generation_seconds_batch": 16.03667499497533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.073120594024658, "target_z": 1.7890623836234505, "ce": 6.356205940246582, "genuineness": 4.218649387359619, "identity": 0.5980132818222046, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8d551a5d6572a961d7db0946181829576e3549da645822ccbe5f7335b47f86bb.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "8dc052ffcdc2facf40415e460782c40316a197e05dfa693d40aabb9b06b8ddb3", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d7817c6a8eca624bd152920d942ff053e5b6eb80fbb0bc8981b271526c59ed11", "generation_seconds_batch": 16.564654507907107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.1020402908325195, "genuineness": 4.174716949462891, "identity": 0.6471635699272156, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8dc052ffcdc2facf40415e460782c40316a197e05dfa693d40aabb9b06b8ddb3.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8e359ff96f9ba426f5b69e5a038f77872a3d2e317887d3dfc8b7099274244f7b", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3d094a7467aa7e15b53bd946a729f84105839f488a651528bb604a67e20e9ce1", "generation_seconds_batch": 16.866492770961486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 6.443574905395508, "genuineness": 4.381826400756836, "identity": 0.5036131739616394, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e359ff96f9ba426f5b69e5a038f77872a3d2e317887d3dfc8b7099274244f7b.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "8e3b0b459dd75a06589dae0de9fe9e96dd82241baa191779affeed0ca7a4248d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "54ab747a64d909148dd2b4a5d19868924d633efc96e627c91fa479f89954ee45", "generation_seconds_batch": 23.593673208961263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.086154937744141, "genuineness": 4.432034492492676, "identity": 0.7505990862846375, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e3b0b459dd75a06589dae0de9fe9e96dd82241baa191779affeed0ca7a4248d.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "8e53c9580acc9370ce7c64b3b6116c0a11fe275a0a04dd896290d654ce30e67c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1157ca4b866dba75fafd584ddbed38046a956f52f04c745aea54083bf82d41e9", "generation_seconds_batch": 22.558813657960854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.2890625018472468, "ce": 6.308549880981445, "genuineness": 4.545003414154053, "identity": 0.7347157001495361, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e53c9580acc9370ce7c64b3b6116c0a11fe275a0a04dd896290d654ce30e67c.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "8e5e028a6e654cb844b04ffc4b4d236a74a5710b452a53b1daee4d8d5c8540bf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5b6a4919d2c86b8cf3aa31c7a675abeaab3825a9ab8d1d0d97ca0dad4c447c15", "generation_seconds_batch": 21.838809613022022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.316492557525635, "genuineness": 4.168440818786621, "identity": 0.48256203532218933, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e5e028a6e654cb844b04ffc4b4d236a74a5710b452a53b1daee4d8d5c8540bf.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "8e60c8707987e9f74dd2aa62116ef646a7d966815e00937f135816baa00bf7cb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f7499438ebe286325d289f1dc5e2bcad89a45d7965a704389c5ba49a2fb3e022", "generation_seconds_batch": 15.133732657996006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 1.179687520319715, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.7706483602523804, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e60c8707987e9f74dd2aa62116ef646a7d966815e00937f135816baa00bf7cb.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": "8e79c85e33b4c2d5cba404be8abca11bd5b9b4412543cbff8a4b77a01ed9dc8e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95a725d6f7557a5d2029b232fa2b3cdf680307b650de48f1c902eb4c6fe9480b", "generation_seconds_batch": 21.211452270043083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.284721851348877, "genuineness": 4.193545341491699, "identity": 0.7438549995422363, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e79c85e33b4c2d5cba404be8abca11bd5b9b4412543cbff8a4b77a01ed9dc8e.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8e8e2c0b69226755cd79813f6b54a1147954bdf82288b1c17c018170287bb352", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "98ce35e1173662de9e4038e3b3913a35e96b33712b54c4559ce402d5040fd7f6", "generation_seconds_batch": 15.653577949036844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.442910671234131, "target_z": 1.1640623540675015, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.2519153654575348, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8e8e2c0b69226755cd79813f6b54a1147954bdf82288b1c17c018170287bb352.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "8f1da7b20091dff2565e0de49f3a1cb82634cfaa954a460248b76ec91321c72e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "345dab2dd99e1fd72055c17fd683a4025734b26d2c0f49aaeb6ab5aa841f4520", "generation_seconds_batch": 15.689369329018518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.038498878479004, "genuineness": 4.406930446624756, "identity": 0.7035807371139526, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8f1da7b20091dff2565e0de49f3a1cb82634cfaa954a460248b76ec91321c72e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "8f3beea745fc222ec1f6cf9d5a845d4cb5f74ffc6382ea90dcfb8dafc951f58b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3fc4d8762869018fdb5716ade1793197e6105e8fa3d4d20ea4ac753b61ea26e4", "generation_seconds_batch": 22.105579582974315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 1.1015626348490175, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.7056099772453308, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8f3beea745fc222ec1f6cf9d5a845d4cb5f74ffc6382ea90dcfb8dafc951f58b.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": "8fb7f8343a5fe2583fe9e27749742fa95c8783ce9446360c84115ec9d99a7937", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5df56cea464f68870a4153d4ab15d4acb7d8181c8bc59a847b0736cc9be3deaa", "generation_seconds_batch": 20.95910456997808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.260893821716309, "genuineness": 4.042920112609863, "identity": 0.7159180045127869, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8fb7f8343a5fe2583fe9e27749742fa95c8783ce9446360c84115ec9d99a7937.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "8fc47b2b86936aa8871c8fc5f479ff16f056dca5684b8db90140d64c5711ac8a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a8cf35c4e75e95c8c9086bd7fc8ede32a355ed2619b6b5efb5ba5a756f4f91ee", "generation_seconds_batch": 24.349841281073168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2065820693969727, "target_z": 0.9296874612078169, "ce": 6.316492557525635, "genuineness": 4.482243061065674, "identity": 0.279468834400177, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8fc47b2b86936aa8871c8fc5f479ff16f056dca5684b8db90140d64c5711ac8a.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "8fd5d7a3d268b842e7348c8ff832ad43a9c97e79bf82783aacf565fc7ab54b47", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "55f54313bcda428db7737a009ccbc44279ba528dfdba97deca023de6dcd06e58", "generation_seconds_batch": 21.05324939393904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.252951145172119, "genuineness": 4.231201171875, "identity": 0.6631811857223511, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8fd5d7a3d268b842e7348c8ff832ad43a9c97e79bf82783aacf565fc7ab54b47.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "8fe472646872d2582dfeb575ecdb5b44c855bddfadac0ae8f2a51fbe06df00fb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9c9b9488338a3822c956bff8d743a330c18530cb13ba995d995b5de1bb676f90", "generation_seconds_batch": 21.86483442108147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.1894097328186035, "genuineness": 4.15588903427124, "identity": 0.5281286239624023, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8fe472646872d2582dfeb575ecdb5b44c855bddfadac0ae8f2a51fbe06df00fb.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": "8fe8d4366df0c1b05b041ccec2f6a144f9de5f61f65520f84bd424ca76fe9857", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "bf95a956355126b548d876d7f48b66ccd99ac5f3b8b12714c4bd017fce347abc", "generation_seconds_batch": 18.172833193908446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.475345611572266, "genuineness": 4.369274139404297, "identity": 0.30803626775741577, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/8fe8d4366df0c1b05b041ccec2f6a144f9de5f61f65520f84bd424ca76fe9857.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "90b9bd8e5ca66e06a6c5f263184f8e4d8cd8c0d38e380be3c04adc4e780ad0c9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "07b00a956cd2b5dc8f904871aadb308a727c30eb0c9f127efbef14e1e87d9064", "generation_seconds_batch": 16.518168410053477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.445312509236234, "ce": 6.356205940246582, "genuineness": 4.337893962860107, "identity": 0.33286362886428833, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/90b9bd8e5ca66e06a6c5f263184f8e4d8cd8c0d38e380be3c04adc4e780ad0c9.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "911a74e2da5b2c8e766cff6319258506b3e009733e2038dfd368833447df909d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2dc533015ecc496ec78fda795842d64f79014faa09951770bd5ed4beb49c57d8", "generation_seconds_batch": 21.7227265490219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.093750169946707, "ce": 6.372091293334961, "genuineness": 4.118232727050781, "identity": 0.30616676807403564, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/911a74e2da5b2c8e766cff6319258506b3e009733e2038dfd368833447df909d.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "91b61436dc33afec0ef18234685d84c2a7269e352fb42fb7b6262d70d35ba6d6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f6a35fe26af71ef67c005e496644dc1f5b9fdd0b5381b7de20cad0ca785bb229", "generation_seconds_batch": 20.372033560066484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1278059482574463, "target_z": 0.8515625757371195, "ce": 6.149696350097656, "genuineness": 4.20609712600708, "identity": 0.40157368779182434, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/91b61436dc33afec0ef18234685d84c2a7269e352fb42fb7b6262d70d35ba6d6.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "91f0708ca4b7f187623acf7d12257345ab39e2c41c5f630ce2521c8ba8a33e5a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bbbf09f3a85df9cb7dd1d6fa6625fb21f7f6535306d2d7ee4cf6568bb436a7a2", "generation_seconds_batch": 16.223388605983928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.348263263702393, "genuineness": 4.632867813110352, "identity": 0.8025332689285278, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/91f0708ca4b7f187623acf7d12257345ab39e2c41c5f630ce2521c8ba8a33e5a.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "91f254f44f1dfe24e15f9a232d922f238e650f48b092e289ffe6c4b36fb80d7d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "90f608f0b822654736a6edb6dfa6e489c24df585de87d9ab45bf38d21ed84729", "generation_seconds_batch": 20.49214577092789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.8283469080924988, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/91f254f44f1dfe24e15f9a232d922f238e650f48b092e289ffe6c4b36fb80d7d.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "91f877e63a08b527db22f4b03bb0ba68d0d91eaef569332d2bb5a460c3330326", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d8a5907f45097dc8378abe6903c10e17a1f69f70e55bbf0a6c117486c1af1509", "generation_seconds_batch": 22.170389643055387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.804687549875664, "ce": 6.070269584655762, "genuineness": 4.256305694580078, "identity": 0.6946929693222046, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/91f877e63a08b527db22f4b03bb0ba68d0d91eaef569332d2bb5a460c3330326.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": "9211d6308f39a433d1fbc8c75654b835d8593264e36f8b1b23985916eb67f9dc", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61937d7b02ed07c50c815bc5f85539acc886123e9f781bd290883ba5ed464f9a", "generation_seconds_batch": 15.92403449199628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.073120594024658, "target_z": 1.7890623836234505, "ce": 6.157639026641846, "genuineness": 4.1998209953308105, "identity": 0.6360634565353394, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9211d6308f39a433d1fbc8c75654b835d8593264e36f8b1b23985916eb67f9dc.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "9288fb4fa22a2fd0611267ccb294333f53f0963c67a1caeaa0eb7fbe10ef84f2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d38a5c8ac3ee2b2b68706309c61509dcd8b084e488fbf7b305b18b82bd373738", "generation_seconds_batch": 15.23185336496681, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.133810997009277, "genuineness": 3.9425034523010254, "identity": 0.44822415709495544, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9288fb4fa22a2fd0611267ccb294333f53f0963c67a1caeaa0eb7fbe10ef84f2.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "92a92f8439169eb4809220b452e757d93a6b92d9540a0b36ab1e1bbb8c03ef4a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "32b5733dc0496257a5d1c85bdbf586b89b08b420da2b18166b1b3a1f3d4b8197", "generation_seconds_batch": 22.158739043981768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0608460903167725, "target_z": 0.7851562693960916, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.16061262786388397, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/92a92f8439169eb4809220b452e757d93a6b92d9540a0b36ab1e1bbb8c03ef4a.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "9341516d3befda801bd7994ddb79fdfe1ca509f47a85492ccedff3fddb692554", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e1a960ec400cda9893fc223f6c4759d28eb7b90ce7c5c0945e0b840d05a75d69", "generation_seconds_batch": 21.09964662406128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6166768074035645, "target_z": 2.3281250628063916, "ce": 5.967014789581299, "genuineness": 3.8295347690582275, "identity": 0.6389162540435791, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9341516d3befda801bd7994ddb79fdfe1ca509f47a85492ccedff3fddb692554.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "93c83b50e7051c10e0521cd0e8dd7802827a34eb24a757748632efd390f7c99b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6fcf45bf171ee37ff866dae609d76f8705c21acd17803350073ea509df5b38ff", "generation_seconds_batch": 22.006949549075216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.4687501403907581, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7048965692520142, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/93c83b50e7051c10e0521cd0e8dd7802827a34eb24a757748632efd390f7c99b.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "93da6e33ec7099b99b7dabcf2ab15104a2a4b94611b9cca11951bb601bbf55b2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "df1c1b114592a392cfb0a8abb0c87d7ad96fc106e25646878dd06f3ef24d5cfe", "generation_seconds_batch": 22.16830753395334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0253968238830566, "target_z": 0.7500000591118982, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.313010573387146, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/93da6e33ec7099b99b7dabcf2ab15104a2a4b94611b9cca11951bb601bbf55b2.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "9413ea4a1e8b535cabe4de144d12753a22b3b4640cc2e498141ae1f3cc19d98a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b47d2b97b8b48327130a8792b7dcca29c202f6fddc24e6c238b3e9be21bbfed9", "generation_seconds_batch": 16.465189505950548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6949949264526367, "target_z": 1.4140626496269921, "ce": 6.181467056274414, "genuineness": 4.532451152801514, "identity": 0.6920245885848999, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9413ea4a1e8b535cabe4de144d12753a22b3b4640cc2e498141ae1f3cc19d98a.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "94b3c73e61dcae3332b3fc28f416f4978d6a2a320e97e5fa26d9e7bccaee8c61", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b0f4c8bc6c4b231164425309405c47294f7c115dc1f1526f30571aa89bca017b", "generation_seconds_batch": 24.07217539602425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.6966500282287598, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/94b3c73e61dcae3332b3fc28f416f4978d6a2a320e97e5fa26d9e7bccaee8c61.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "94ee505fe4ee49587aa5a217a8408d92f6fd9c19891f170c547181a2981935e9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9b16ec42bee54035df0aaf28ab048e1f1cb2df92d04f2d24535d94cfec26f243", "generation_seconds_batch": 22.840981137007475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.3984374833747786, "ce": 5.967014789581299, "genuineness": 4.425758361816406, "identity": 0.7149207592010498, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/94ee505fe4ee49587aa5a217a8408d92f6fd9c19891f170c547181a2981935e9.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "9552e7878674a7b6e323a0791e7017aacd8675822e69db291127a3e190d55dbc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d82430bcd68d39fa2c5f56ac523264f4afb1d0a2910667957869ce76fc24eeba", "generation_seconds_batch": 20.15725921303965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.006728172302246, "genuineness": 4.494794845581055, "identity": 0.7153925895690918, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9552e7878674a7b6e323a0791e7017aacd8675822e69db291127a3e190d55dbc.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": "95990c67b960a3395f000e5c551f6c846894a64a08f81d4c5c329122ce5ecf98", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2844febae4e5e09ad9e6e8f80ad4b2561c232f8d9c1fe0b9adde6ea92a89b63", "generation_seconds_batch": 15.685634549008682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.017518997192383, "target_z": 0.7421873577619952, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.8070827722549438, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/95990c67b960a3395f000e5c551f6c846894a64a08f81d4c5c329122ce5ecf98.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "9692e503e52ebb2164553ccf02a808c114dc52bc00fbf409e0dcc5b95c3e1711", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7364a87526521106d750b991de350c45ef1fba75798b70fd08c8682b128c3756", "generation_seconds_batch": 21.857127454015426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 1.1328124944582596, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.3794364631175995, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9692e503e52ebb2164553ccf02a808c114dc52bc00fbf409e0dcc5b95c3e1711.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "96f18f7af715eab49c738bc58c006bca26ce26e24da9398719bdcb31c35dc7fa", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2d7a105892f680e83b756a1d8f4d78b13401b42eef4c997621e949cf47c027d5", "generation_seconds_batch": 22.73400146397762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.774229049682617, "target_z": 2.4843748337477867, "ce": 6.292664527893066, "genuineness": 4.21237325668335, "identity": 0.636016845703125, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/96f18f7af715eab49c738bc58c006bca26ce26e24da9398719bdcb31c35dc7fa.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "972acdf9d7af44afb12a8719cf55a637ca7c9cabff82383c1f9b9ce84da3c0b1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "88bb7c64fb3d866b56c2de454b9d2252784ef50802f42dd8f52e0ebdee8a6b6b", "generation_seconds_batch": 23.12865208403673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.078212261199951, "genuineness": 4.281409740447998, "identity": 0.6757786870002747, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/972acdf9d7af44afb12a8719cf55a637ca7c9cabff82383c1f9b9ce84da3c0b1.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": "9768d56e1d89a3738908c249df6e8376d5cc54399b01e9d46a72d63babc13e76", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f65a68ade412ff7675cdf3c619c4a29d1763dbd17f058c597e912b7d8b0f3d8", "generation_seconds_batch": 21.26246481097769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.8740265369415283, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9768d56e1d89a3738908c249df6e8376d5cc54399b01e9d46a72d63babc13e76.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "97e89c12fd50a6fe518673df9f12507d39979374b732ac91a835c4327af13dbd", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "545b08aecd4e7d624932ab40c7cfc628d64ee1fddfa83d4cee29bc450db97689", "generation_seconds_batch": 22.450672763050534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.205295085906982, "genuineness": 4.143336772918701, "identity": 0.2387726604938507, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/97e89c12fd50a6fe518673df9f12507d39979374b732ac91a835c4327af13dbd.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "9805a6507728ac9852f0678dc335dc41b4e8c859ecfcb2b77a34976408012118", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "afb42e35e03557329c4c3986efee131a65514ea402a1a98e707ca021f9634835", "generation_seconds_batch": 22.691166303004138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0214579105377197, "target_z": 0.7460937084369467, "ce": 5.764476776123047, "genuineness": 4.444586753845215, "identity": 0.7642871141433716, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9805a6507728ac9852f0678dc335dc41b4e8c859ecfcb2b77a34976408012118.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "984a52b214c9bb676235b5e495ff5579520434db592a16ba05482c05685c182c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ce196fd872790c08b3db6b6feb47d6cfcf08cd90cfc8913f9dba80f9e76bb21b", "generation_seconds_batch": 15.929894828936085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0923566818237305, "target_z": 0.816406365452926, "ce": 5.998785495758057, "genuineness": 4.243753433227539, "identity": 0.7392010688781738, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/984a52b214c9bb676235b5e495ff5579520434db592a16ba05482c05685c182c.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "986c57d28e5bfd126389eb0880cf96001c05e1160388e343289a736fab77a3d2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2eb1503d5e0ee982321e41d1cf594511004c8d49948d3b39df41f8ead14a5787", "generation_seconds_batch": 16.736632735002786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.3641486167907715, "genuineness": 4.494794845581055, "identity": 0.4994610846042633, "hypothesis": "I found the little bock. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/986c57d28e5bfd126389eb0880cf96001c05e1160388e343289a736fab77a3d2.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": "98ac3096adf0d1e9f4f6c3042899c6b80ef9627877a2d9decfb78d2d86e67cf6", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e38271d20ab2d5cbe86601d76b8419afa9b71534e847475bb437fc84cec6456d", "generation_seconds_batch": 21.211452270043083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.252951145172119, "genuineness": 4.319066047668457, "identity": 0.7168207168579102, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/98ac3096adf0d1e9f4f6c3042899c6b80ef9627877a2d9decfb78d2d86e67cf6.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "99a0566d636382aa42995ce4b4b165d3798f90768b00b0a0ecd839df4426f514", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "604fe369e442fd6881dae28c7ed56e674d110f4beb472d756d4b889dec0bc772", "generation_seconds_batch": 16.05603813799098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.962833881378174, "target_z": 1.6796874020959187, "ce": 6.197352409362793, "genuineness": 3.9801597595214844, "identity": 0.36015087366104126, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/99a0566d636382aa42995ce4b4b165d3798f90768b00b0a0ecd839df4426f514.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f125dd68f09ca352b4d9c326937b6f57a884105f6f7beaa7a3c4d8a1c9e2853", "generation_seconds_batch": 15.534635560004972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.1562501256127835, "ce": 6.372091293334961, "genuineness": 4.432034492492676, "identity": 0.6152548789978027, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "99f72553eaa588d458a282827826f8c54d54fad4e125a6c70bb390c205476e0b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9f9d9eb4251e22a8a666f9a9a456d7eb7be41859965124fa3ba470b99600e1dc", "generation_seconds_batch": 22.635253492975608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.022613525390625, "genuineness": 4.319066047668457, "identity": 0.7474931478500366, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/99f72553eaa588d458a282827826f8c54d54fad4e125a6c70bb390c205476e0b.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "9a2d3af1d322b6978b4e18cd528dfa0466723724766a9dead15085d1f0353e52", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1356a4a4758281624eb614b1a039526e2565857c9ed4fb2c68c6497d825d5dbc", "generation_seconds_batch": 21.71362113195937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049029588699341, "target_z": 0.7734374538188296, "ce": 6.324435234069824, "genuineness": 4.494794845581055, "identity": 0.46185627579689026, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9a2d3af1d322b6978b4e18cd528dfa0466723724766a9dead15085d1f0353e52.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "9ace9b039bf63f319a8bb1690fbe43024cc02293c1cd243b6a58f17aca34c79c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b4de50c273af5d7ee65c4aab8ab6415567f6cdc9296aafc46fa740bbe85f45a", "generation_seconds_batch": 22.969700744957663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.6480343341827393, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9ace9b039bf63f319a8bb1690fbe43024cc02293c1cd243b6a58f17aca34c79c.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "9adb80934bd8205fc9ac38e91357b321f8431586788a61a8e8340fbecde6d242", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f19dfbb189b234a3397193554e83df95aa607be6cb55e0144f5efbb2d2c91643", "generation_seconds_batch": 16.179021560004912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.5949094295501709, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9adb80934bd8205fc9ac38e91357b321f8431586788a61a8e8340fbecde6d242.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2de033ef42a6f7e192a2909fdfa874647c3e64d6c0172d0bc1cf705961e252e5", "generation_seconds_batch": 16.552686794893816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1278059482574463, "target_z": 0.8515625757371195, "ce": 6.252951145172119, "genuineness": 4.582659721374512, "identity": 0.8471251726150513, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "9c13242f6810227cff8f0eb8adbf6d70090157350e8b9f95be657635d04791b1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "deac9d65f2c7d5ac68b3d81c479bfacee0cb5e82a5a309f5c1e48580e966e277", "generation_seconds_batch": 22.28865651297383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.6076468229293823, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9c13242f6810227cff8f0eb8adbf6d70090157350e8b9f95be657635d04791b1.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "9c9475838c817f49f0e66b3f8f5882f9f3f3435cd6501128e2f8698364f6ecc0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d442f3bbcb6d7b98ba9f7c194165082a82926cd7f67db0faa89343a995699356", "generation_seconds_batch": 15.928393853013404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.109375099751328, "ce": 5.95112943649292, "genuineness": 3.848362922668457, "identity": 0.7097166776657104, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9c9475838c817f49f0e66b3f8f5882f9f3f3435cd6501128e2f8698364f6ecc0.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81266ebbc87a85ecb0f1de4a519b27731ee3cbea38bccd2422ad2914c271f15e", "generation_seconds_batch": 15.745734230964445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3641345500946045, "target_z": 1.0859374685968042, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.6513363122940063, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "9d77f415f0abf2fd1bf3799f081d5d8fc73a4266c1b5c8bdb72dec1f4d68902e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c7c1551a5b4173b2a89f225ae236517f1a7a9ce349d2869da4907996f24ad4d1", "generation_seconds_batch": 21.73912115301937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.062326908111572, "genuineness": 4.105680465698242, "identity": 0.7406271696090698, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9d77f415f0abf2fd1bf3799f081d5d8fc73a4266c1b5c8bdb72dec1f4d68902e.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "9e2885bf07330e01cf7bdf0ec502af3ec8fc836f5a5f35052e9beb820f827d60", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "254b2a566b49a1a565828ee123f1c510e9521eac051514f2fd885cf661544985", "generation_seconds_batch": 21.459020851063542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.561126708984375, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9e2885bf07330e01cf7bdf0ec502af3ec8fc836f5a5f35052e9beb820f827d60.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "9ea497d1e099bc286894d8704fa23c1f34a6320c0a35de7fd62ef1bd6cd33633", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f9ceb010698294ecc86cf0a6c6b5442f7a5d23fc9713498fb9aca7bed5598dec", "generation_seconds_batch": 16.775594857987016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.2968752031971498, "ce": 5.959072113037109, "genuineness": 3.653805732727051, "identity": 0.8382423520088196, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9ea497d1e099bc286894d8704fa23c1f34a6320c0a35de7fd62ef1bd6cd33633.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "9ef7cdd82dbcacd5ce805371c928cfe047ae3d48dc7185293b31dd82f0e56012", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "92e7617a75d13f8911bd01218be1223ef7ec570ae360b1427e9b5b2ad7b80396", "generation_seconds_batch": 16.10289249906782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.340320587158203, "genuineness": 4.344170093536377, "identity": 0.5283800363540649, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9ef7cdd82dbcacd5ce805371c928cfe047ae3d48dc7185293b31dd82f0e56012.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "9f0443879565215c30c20416179d903d72730b60c6f0415587943f7f2436c25a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21affd45047bfe777dc7194f6d0a0e28549273150f9083eac3d92707781d0cf0", "generation_seconds_batch": 23.85829269804526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.756873369216919, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9f0443879565215c30c20416179d903d72730b60c6f0415587943f7f2436c25a.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "9f0ec73ef4f76de3863f26acf8d48d4941e58ee17df0328518f3084fef2ad48c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d446f2539f5b229bec21c54ad3e89ac77044da59bbe331c95b9d629d175b588b", "generation_seconds_batch": 21.7470355440164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7851293087005615, "target_z": 0.511718815577262, "ce": 6.038498878479004, "genuineness": 4.833701133728027, "identity": 0.7902495861053467, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9f0ec73ef4f76de3863f26acf8d48d4941e58ee17df0328518f3084fef2ad48c.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "9f3dbaca21befcbd5eebb96244a260229b933280cb6ed7ede53456ac2ff2ddac", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5392f4ac12006d1af5183518a78be8b6e9ab0855d8e669bbdb7feaed1a53c89e", "generation_seconds_batch": 16.387265507946722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2183985710144043, "target_z": 0.9414062767850788, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.7923159003257751, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9f3dbaca21befcbd5eebb96244a260229b933280cb6ed7ede53456ac2ff2ddac.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "9fb1f0ad45301f65b44a5947f93882233cc2a2e14d4773a6cb12fa8a4f126fc5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "68dfe51f4a2c6c08cbb7a9bad09ddf8f6b2903be40d95319cbc77892b8f411c6", "generation_seconds_batch": 21.916390803991817, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.062326908111572, "genuineness": 4.337893962860107, "identity": 0.611056923866272, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/9fb1f0ad45301f65b44a5947f93882233cc2a2e14d4773a6cb12fa8a4f126fc5.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": "a03d352d9818338e67da4513c7abdae560a64f3b62b867a6318bf661fa515f01", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e72a322b3b03abd2c3acd72c0b69d0759510757606584ab6b8664d8c4313962c", "generation_seconds_batch": 15.913670809008181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9374999261101273, "ce": 6.38003396987915, "genuineness": 3.8797430992126465, "identity": 0.11696913838386536, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a03d352d9818338e67da4513c7abdae560a64f3b62b867a6318bf661fa515f01.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": "a0932f46c62d064c677948fcd2862c2e21d87d844d3c1df4ed6beb858ea731bb", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f565eaf205562ca26113e2b4eb0ddff94607b0316d62f5e83f77ec2ebfe12527", "generation_seconds_batch": 22.77096488000825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.9843749519715828, "ce": 6.157639026641846, "genuineness": 4.607763767242432, "identity": 0.6259438991546631, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a0932f46c62d064c677948fcd2862c2e21d87d844d3c1df4ed6beb858ea731bb.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "a11b8fddbb9059943bf2a7e5ccfdeac5b2db5d3b249bb2dde617460cd76eb9a5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9f595354e2986d5d836151c8d4d9fff13e35805911ea8d66996a5d50870ca780", "generation_seconds_batch": 17.23408327891957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.1562501256127835, "ce": 6.181467056274414, "genuineness": 4.15588903427124, "identity": 0.712944507598877, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a11b8fddbb9059943bf2a7e5ccfdeac5b2db5d3b249bb2dde617460cd76eb9a5.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "a1f4905b2649bd2dad49ac73cccebefb59f1ca5111ae1f6e8f136fc906f7be72", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "59d72289d8ab7dc67dbbacafc67ff76ed36e88f60c14c4479885c6683d9a1b82", "generation_seconds_batch": 20.15725921303965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.117925643920898, "genuineness": 4.6203155517578125, "identity": 0.7483751177787781, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a1f4905b2649bd2dad49ac73cccebefb59f1ca5111ae1f6e8f136fc906f7be72.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "a27b43e6a97c8e3bf89bb0aee980cb09e37bbac37c67c94bae602f98cbe52cf1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "dd6a7445a670aa523594df41e72e98a621814741e151af5c8e1a77c2ebe87ed9", "generation_seconds_batch": 20.31884474004619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.165581703186035, "genuineness": 4.1998209953308105, "identity": 0.748272180557251, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a27b43e6a97c8e3bf89bb0aee980cb09e37bbac37c67c94bae602f98cbe52cf1.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "a2a18a62005f20cf4c3458110572074f582b385a95f212a44c01555c48bd300b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9b003a9ee91be083b78a7178c7f0fd4b4b48ca4cf89f2ac07fc1a414eff69a36", "generation_seconds_batch": 16.517090607085265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6162185668945312, "target_z": 1.3359375277087022, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.7852414846420288, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a2a18a62005f20cf4c3458110572074f582b385a95f212a44c01555c48bd300b.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "a2b73687ffa8a3d08c6850b1fa40454d8b052708014cf1c9d83b568a48b89864", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0acbd427c93f31f9f2b6361ab8c1f09938fe240813d17c6d31cb750ff1d33dd5", "generation_seconds_batch": 21.969054025015794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.221180438995361, "genuineness": 4.13078498840332, "identity": 0.6922810673713684, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a2b73687ffa8a3d08c6850b1fa40454d8b052708014cf1c9d83b568a48b89864.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "a2c53abdef195eb40c735444cb9799b57a037f910ce5880911942a1df2f1d661", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "604b87da93cf9d1778cd82c91923262d58a61fcf39c85a83a842894cd11c91db", "generation_seconds_batch": 16.502552989055403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537442207336426, "target_z": 1.2578124057904123, "ce": 6.308549880981445, "genuineness": 4.325342178344727, "identity": 0.194802388548851, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a2c53abdef195eb40c735444cb9799b57a037f910ce5880911942a1df2f1d661.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a2e034774762b86d2863643d4ce2c0f279545f29020e3907e349a62be569bf0d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6abcae8a290ccc99b164a550698797ca74710b2604efe8fac7c3d4df9c57ac05", "generation_seconds_batch": 16.710920891026035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 5.847874641418457, "genuineness": 3.8420867919921875, "identity": 0.5495569705963135, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a2e034774762b86d2863643d4ce2c0f279545f29020e3907e349a62be569bf0d.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "a2ee186048c451633d3d23064ea3838b5d8a4d8c61ce0464edc04d11dcbfa818", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3d0d44847f028f91005b7bbd38f178a55c445edc03c180a86c6122bac99975ce", "generation_seconds_batch": 23.135817307047546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.197352409362793, "genuineness": 4.18726921081543, "identity": 0.36881041526794434, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a2ee186048c451633d3d23064ea3838b5d8a4d8c61ce0464edc04d11dcbfa818.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "a318b992e055a1243f2af659a4f7353af116dc6fc15d2551c644ff518c129473", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "590716cf0e548ee83768fd5780e9e705480a0a49616dca1ca6ca559fa36eff79", "generation_seconds_batch": 13.660180047038011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.6639383435249329, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a318b992e055a1243f2af659a4f7353af116dc6fc15d2551c644ff518c129473.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": "a40f961dbeab0f27c6590a7e09a3e3431ca61fb2c7dda5b5a2d6b137564a11c9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ad93a2822d4ee2db2efefadb7374f410b5c9dc05bace443c088e75382e58b09", "generation_seconds_batch": 22.780558011028916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.727177083492279, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a40f961dbeab0f27c6590a7e09a3e3431ca61fb2c7dda5b5a2d6b137564a11c9.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "a44b61a1ec644ae233fbd8aaac7eb1a540e2cfef61dcc9d6f9ad388dc061dfcf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a97bc518b4d37583a60ff7ad823e64f49d350c736d0756cb0f4a7376959fca66", "generation_seconds_batch": 16.835587586974725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9938862323760986, "target_z": 0.7187499630550637, "ce": 6.316492557525635, "genuineness": 4.519899368286133, "identity": 0.30301451683044434, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a44b61a1ec644ae233fbd8aaac7eb1a540e2cfef61dcc9d6f9ad388dc061dfcf.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a47ac8a8d7367d743e1bc1a7823c5a75315d94bb3830ac54b0c21e46745e7f9c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8ee3f41ab5c625193d0265abdb9695ec2e05d74bd7522cabc2221715f0083f47", "generation_seconds_batch": 22.973399496055208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.600463390350342, "target_z": 1.3203125979040813, "ce": 6.0146708488464355, "genuineness": 4.2751336097717285, "identity": 0.6408636569976807, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a47ac8a8d7367d743e1bc1a7823c5a75315d94bb3830ac54b0c21e46745e7f9c.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": "a4e4a8b982f62c7b16b0ca12b60286e9a132a6c8e6bd9af4f37d60987fe59b60", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2934c2b39461637c52fd9944d139370bb2a0648dc281da8cdf5fa34747f0224f", "generation_seconds_batch": 15.614241737057455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.789984703063965, "target_z": 2.5000002364475926, "ce": 6.292664527893066, "genuineness": 3.898571252822876, "identity": 0.5827847719192505, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a4e4a8b982f62c7b16b0ca12b60286e9a132a6c8e6bd9af4f37d60987fe59b60.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a4ece27bef22a4ae98853aca345b0c8aa6d7a0d0f4df54706b9d554aa813f89c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "90b9582a3b5d33f3fd9ff2b7b95a91a77014d6e8a55013c03477a082099aad5b", "generation_seconds_batch": 20.70256081793923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.632431983947754, "target_z": 2.343749992611013, "ce": 6.0305562019348145, "genuineness": 3.6977379322052, "identity": 0.6246644258499146, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a4ece27bef22a4ae98853aca345b0c8aa6d7a0d0f4df54706b9d554aa813f89c.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "a50189b7ae4f676c0c331b5cc99fa09f908780a66baf23875e199afe41b6ff0e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b961799ccd76a8d15d60555beaa4c923833da563159da707e8d644d6986f7aab", "generation_seconds_batch": 16.465189505950548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.23706579208374, "genuineness": 4.168440818786621, "identity": 0.7106443047523499, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a50189b7ae4f676c0c331b5cc99fa09f908780a66baf23875e199afe41b6ff0e.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "a523e550664e73217515c0e9569500ca6d8cbaa7049c25679b285065d162df2d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d814eca451eab9fa1706b67e31ceab5f5f8f0f8c024b1889761063e9ea8dec6e", "generation_seconds_batch": 15.442915622028522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.403861999511719, "genuineness": 4.369274139404297, "identity": 0.3103645145893097, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a523e550664e73217515c0e9569500ca6d8cbaa7049c25679b285065d162df2d.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "a5c7f5f98f86ac0114e6c49e1bf897271c6f1a151b7416abc9acd848297b4fba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b2aa4bba5020973e3e231f224453fb855ddcc5d7e2b3dd1ac7c4e438805bf1c9", "generation_seconds_batch": 21.958985276985914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9269264936447144, "target_z": 0.652343774937832, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.723258376121521, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a5c7f5f98f86ac0114e6c49e1bf897271c6f1a151b7416abc9acd848297b4fba.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "a5f500d87940f5b1c80c10d92d5f9e438cecc6d28297108efe2f2a5554e33205", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2e1aadda0a574c34669aa9cfd7e479e30a2ff62a5926aa2d42bfee60bb45ab31", "generation_seconds_batch": 15.83446621207986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.1020402908325195, "genuineness": 4.369274139404297, "identity": 0.726917028427124, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que se ha hecho.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a5f500d87940f5b1c80c10d92d5f9e438cecc6d28297108efe2f2a5554e33205.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "a67a2eaf8e3e2cea8d702cb5a8eccb5608a6d9532b5ba30f5a3ab996c4d2beee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0c3c77c926614199dee8c06963e01191c579f6ceb2b67a27723b89a1b87206dc", "generation_seconds_batch": 23.12865208403673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.213237762451172, "genuineness": 4.21237325668335, "identity": 0.6655263900756836, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a67a2eaf8e3e2cea8d702cb5a8eccb5608a6d9532b5ba30f5a3ab996c4d2beee.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "a6fd3bc0a83c5cdfb73f5a15fd1303cb9ca632f0d91d290d25b7c5a25f83aedc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "10693263b7831b34ecfa5028857a089e847922ab5f0a80db80c268ec90cad6e0", "generation_seconds_batch": 24.388659139047377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.3641486167907715, "genuineness": 4.256305694580078, "identity": 0.4964071214199066, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a6fd3bc0a83c5cdfb73f5a15fd1303cb9ca632f0d91d290d25b7c5a25f83aedc.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": "a71db16a72d20da5452f7068bc27c34f3d17582166770d0f2deea2a067646875", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "923fd269e5299dab58ba3c9aa1dd4d5bfd2b6f96abb124c7d6126ce076340cdf", "generation_seconds_batch": 14.609708734904416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773771047592163, "target_z": 1.4921875350976894, "ce": 6.070269584655762, "genuineness": 4.268857479095459, "identity": 0.7581489086151123, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a71db16a72d20da5452f7068bc27c34f3d17582166770d0f2deea2a067646875.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "a74f45f8af13fa4e5b058e23a833ee9cb891d362ec3af184bb42115834978c29", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a23033404164c75d5dda3833beb65af1a3c4427f6c1f8c3250b523739b3436cc", "generation_seconds_batch": 15.81392538396176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.093750169946707, "ce": 6.1894097328186035, "genuineness": 3.9299514293670654, "identity": 0.7822756767272949, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a74f45f8af13fa4e5b058e23a833ee9cb891d362ec3af184bb42115834978c29.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "a74fb091e472f17e0611f50432635ca6ab49b6695aa8ac6e6ce6c3cbe9b44656", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "fdacfb35112b72e3dcfa6f57940a968a232faeb8b54422f14e28fa54700bc3c6", "generation_seconds_batch": 14.575893678003922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 5.982900142669678, "genuineness": 4.137060642242432, "identity": 0.515790581703186, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a74fb091e472f17e0611f50432635ca6ab49b6695aa8ac6e6ce6c3cbe9b44656.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": "a846e2ae3caf14f974341d9d5268274943aa15bf3d34d04591fd1605012c38c2", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f860a6a6fbd6f360c072f56f2f75a2b9fc744d01fe2872ee5967a7ad200766c4", "generation_seconds_batch": 21.482375354971737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.6706869602203369, "hypothesis": "Quelqu'un déplaçait la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a846e2ae3caf14f974341d9d5268274943aa15bf3d34d04591fd1605012c38c2.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a85b63eccc84244c0fa21ea5cea3f79afba6f08147ce13589f854a65fb3c787d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "72da7e5aee5d9135c546ae202094e558cd0341972853914d40d94b6a60edb981", "generation_seconds_batch": 22.170389643055387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.14062495936057, "ce": 5.95112943649292, "genuineness": 3.9048471450805664, "identity": 0.5724565982818604, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a85b63eccc84244c0fa21ea5cea3f79afba6f08147ce13589f854a65fb3c787d.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "a89f5b6836d9b0eba30bebe296747439d52de4d027b9276110274bc48ac54ecf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d6581e895693c9f909780dbd5bb7f09e427e91097af15b1e440ee0e32dcc5685", "generation_seconds_batch": 22.565222267061472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.202643394470215, "target_z": 0.9257813469804579, "ce": 6.006728172302246, "genuineness": 4.68307638168335, "identity": 0.7493699789047241, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a89f5b6836d9b0eba30bebe296747439d52de4d027b9276110274bc48ac54ecf.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "796b1181141fcf83a885bed29895becd5a55b5f463015e921e0a8cda07adc8df", "generation_seconds_batch": 15.824052672018297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.141753673553467, "genuineness": 4.250029563903809, "identity": 0.7431143522262573, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a8f3a87c739808d4107458763b8f1d2c47965f5a3da8928e998810fab9d1dc84", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "dc90c176c1e39f1d5727c8a1aefee656d28df442de6e43806b12e892e3b63ac4", "generation_seconds_batch": 15.928393853013404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.173524379730225, "genuineness": 4.18099308013916, "identity": 0.5134057998657227, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a8f3a87c739808d4107458763b8f1d2c47965f5a3da8928e998810fab9d1dc84.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "a9324375a5e6ce07bb1adccd3c3422dcd13089dfea3b7b7c2a3c578857754f51", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b8443004fc0c6661e1fbaa2f111978e8bfbc4384f25cc8e4a27065e7e6a0e011", "generation_seconds_batch": 16.223388605983928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.537442207336426, "target_z": 1.2578124057904123, "ce": 6.252951145172119, "genuineness": 4.545003414154053, "identity": 0.7449456453323364, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a9324375a5e6ce07bb1adccd3c3422dcd13089dfea3b7b7c2a3c578857754f51.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c426eb005682300d01aa90737938282883c70fb4431aaaffc5dab90a736184fa", "generation_seconds_batch": 15.863659417955205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.252951145172119, "genuineness": 4.319066047668457, "identity": 0.7726355791091919, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "a94f90c11b50145dbca01959fc97cac956f253d55c89a91358e3f30abef95a9a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3eb4f248b18cec4fc3e2cc9045e5c26c395afe949d8463b502689c7d804abc69", "generation_seconds_batch": 22.691166303004138, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.600921630859375, "target_z": 2.312500133001771, "ce": 5.982900142669678, "genuineness": 3.785602569580078, "identity": 0.5779933929443359, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a94f90c11b50145dbca01959fc97cac956f253d55c89a91358e3f30abef95a9a.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "a95c829e15871a5e2cace268406b5e30110ba677028043c5b74d8f508da93b89", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "07c5f2c2db870ab246922b7587d8c73e2f9eae9fa63794ef17baa8f129ffaffc", "generation_seconds_batch": 24.086284031975083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.054384231567383, "genuineness": 4.18099308013916, "identity": 0.6074105501174927, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a95c829e15871a5e2cace268406b5e30110ba677028043c5b74d8f508da93b89.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "a9fdc3f79fc28a7ce290fe5ac09e67caf2b122b6e9be65bbc16f6f707d3ba604", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b951cf006046c5ed67f643de8b64759d246a84a4d8ee14d30193eb9cd07a56b1", "generation_seconds_batch": 22.285901729017496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.133810997009277, "genuineness": 4.093128204345703, "identity": 0.6521135568618774, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/a9fdc3f79fc28a7ce290fe5ac09e67caf2b122b6e9be65bbc16f6f707d3ba604.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "aa00dc0bd3fac2d1f52e26e8a1d3b72fc7382e535d184c428fbba44b7b28339a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "389fce485bd06f590d8b9fd7ba28c85b636a6ba6af2bb107f05931d0e817e3a5", "generation_seconds_batch": 21.392400279990397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1199283599853516, "target_z": 0.843750110834809, "ce": 6.300607204437256, "genuineness": 4.256305694580078, "identity": 0.3400111198425293, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aa00dc0bd3fac2d1f52e26e8a1d3b72fc7382e535d184c428fbba44b7b28339a.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": "aad47e1ead18bcf3b3b47541452d336bfc20cc3f3d9dfabcf21cecbbd85bc5ec", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57fcb07e8a4b8e1ab799944419bed415b9dddabca93b5b014b23b175f7e70de2", "generation_seconds_batch": 16.07405153394211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4906349182128906, "target_z": 2.203125151474239, "ce": 6.23706579208374, "genuineness": 4.049196243286133, "identity": 0.6130075454711914, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aad47e1ead18bcf3b3b47541452d336bfc20cc3f3d9dfabcf21cecbbd85bc5ec.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "aaf3c51a9f5d8e051460a90900086005af616a2ea19422589b0edaa269e74b15", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "299d2804262856a4538a7087d6c74ba3cf07eeaf987df4034d1ac6bec777bfee", "generation_seconds_batch": 16.162636072025634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 6.070269584655762, "genuineness": 4.086852550506592, "identity": 0.6132272481918335, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aaf3c51a9f5d8e051460a90900086005af616a2ea19422589b0edaa269e74b15.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": "ab08379dda530de09b8c929d0f0ebf19e3a5889be651f0343c6cf9ca5c3adb74", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba9749abbef3acbe8b335f321750723bfb43d0bb93aa174155b9d7eedcff0179", "generation_seconds_batch": 16.216828600037843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.752435564994812, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ab08379dda530de09b8c929d0f0ebf19e3a5889be651f0343c6cf9ca5c3adb74.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "ab31cf014200f9b9af128478c003339c8a6af9146834578815884d8fa8dfa6d8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "57381503bf610c5cff003c61d5073391d74645b3dc193c453cabd23d763f44d5", "generation_seconds_batch": 21.295317998039536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.5390625609591448, "ce": 6.006728172302246, "genuineness": 4.306513786315918, "identity": 0.589461088180542, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ab31cf014200f9b9af128478c003339c8a6af9146834578815884d8fa8dfa6d8.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5818eb3aa32dfb7be9af082d4ad500cf7a253b3696f71ef3d5f3aee93f690007", "generation_seconds_batch": 16.725593436975032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1199283599853516, "target_z": 0.843750110834809, "ce": 6.38797664642334, "genuineness": 4.218649387359619, "identity": 0.2885127663612366, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "abf83ef19da536ac4075ed12bdcbbc0d1753f26a7139144520f499a9ef2bf813", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "de6695c8a9814693aa56c49213c31ab43fc695c9e1474b8933c1cb5d4693d6ef", "generation_seconds_batch": 22.53672899294179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 5.990842819213867, "genuineness": 4.124508857727051, "identity": 0.6898051500320435, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/abf83ef19da536ac4075ed12bdcbbc0d1753f26a7139144520f499a9ef2bf813.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "ac904da743c699250f6e36ea1074e9341503735476f01a72ff1d641b82cfdf72", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "746729678b3ec5fd5215cab4c3dfc4fa3ea96fa998ca815020c7c6101bc53220", "generation_seconds_batch": 21.451275691972114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.3984374833747786, "ce": 6.38003396987915, "genuineness": 4.369274139404297, "identity": 0.6295930743217468, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ac904da743c699250f6e36ea1074e9341503735476f01a72ff1d641b82cfdf72.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "acb2ad749ce9b48507b874c60b03d27a0206828ac94f4e299336c1b330e287d2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "817e703bb6f1aadd19043c001a7e223af01d2fee2a630083ca11485d3f4c6c29", "generation_seconds_batch": 21.838809613022022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.7395918965339661, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/acb2ad749ce9b48507b874c60b03d27a0206828ac94f4e299336c1b330e287d2.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ad30e7698bb597e2ac2da588f7b7befb09c4b7b2c1d2e05fb6d6bfeb9d2c14a4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0549fd603fe57b572845d9fc35ff10438a60cc001bb1c1a46aeaee91d0c4bbfa", "generation_seconds_batch": 21.549633068032563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.498054027557373, "target_z": 1.2187498448312675, "ce": 5.665193557739258, "genuineness": 4.068024158477783, "identity": 0.7732524871826172, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ad30e7698bb597e2ac2da588f7b7befb09c4b7b2c1d2e05fb6d6bfeb9d2c14a4.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ad3e1ccda81d8340bfa8652d1c88f1695e12b925de170bdbac29e902ca280c5a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f56e8e09d5b6d2e3dbdd39e084964bc2fd1a1a3ffea70cce504c0fd0c6f6812b", "generation_seconds_batch": 27.35881480097305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.9999998817762038, "ce": 6.117925643920898, "genuineness": 4.482243061065674, "identity": 0.8243046998977661, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ad3e1ccda81d8340bfa8652d1c88f1695e12b925de170bdbac29e902ca280c5a.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": "ad6d9647d404f8e8a4f574ce0025de8e83ed9c2601d1bc712b97250dff8a1574", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "419deececa92fcf4483c9f0c452f22f38eab7868c8a570b5273d762390bea80a", "generation_seconds_batch": 21.26246481097769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7830244302749634, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ad6d9647d404f8e8a4f574ce0025de8e83ed9c2601d1bc712b97250dff8a1574.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "adb96aaed45a8335f359a29a90f13b4499596122b05ea422d12848fe08285917", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4b5b7f0c37eeefb4d7aed1e5f965da9af4c7d5dc9a7a811a84229a26bed9a732", "generation_seconds_batch": 15.81392538396176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 5.982900142669678, "genuineness": 4.149612903594971, "identity": 0.614046573638916, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/adb96aaed45a8335f359a29a90f13b4499596122b05ea422d12848fe08285917.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74ba0abda78fcb2e9b307fa36e6f857ca49d0df5b06fdebfc604277fb8f34a95", "generation_seconds_batch": 15.010244044009596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.261725425720215, "target_z": 0.9843749519715828, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.3698570728302002, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "adf95589540dd00844f61f67b796ad26d27ff4bebbd81a6dc1b9de6ca52a027c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f51a3f934efdb1bd7841ea8f3060babc727634e340ce5a7c8c1d0ea5ca52fb00", "generation_seconds_batch": 20.202318134950474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3405017852783203, "target_z": 1.0625000738898727, "ce": 6.054384231567383, "genuineness": 4.256305694580078, "identity": 0.7844704985618591, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/adf95589540dd00844f61f67b796ad26d27ff4bebbd81a6dc1b9de6ca52a027c.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "871e72df2b463c33da278158d8fb3eacc6b19615a52c56775bbf12ff426a4813", "generation_seconds_batch": 16.778370628017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7187499630550638, "ce": 6.372091293334961, "genuineness": 3.9676074981689453, "identity": 0.2695930004119873, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "aea4e9d79b5cdc78c1c83c5f931659716a2565bb86004ab20a6919d1af45133b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a31b22c7ba080bd063e869332550bde0b1aa5cb3301282197cab7269819c6eee", "generation_seconds_batch": 20.299461154965684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7536187171936035, "target_z": 0.4804687195204275, "ce": 6.38003396987915, "genuineness": 4.545003414154053, "identity": 0.4604439437389374, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aea4e9d79b5cdc78c1c83c5f931659716a2565bb86004ab20a6919d1af45133b.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "aea84503f0549534f1b0b5ca4409558e27dcdf8b4793fa47b964f8183120fbbb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d3c0ac35096ebb1cc5ad577f97d63583f1568e9123d7ce259c657e8aa01ae997", "generation_seconds_batch": 15.065925286966376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.181467056274414, "genuineness": 4.268857479095459, "identity": 0.6863611936569214, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aea84503f0549534f1b0b5ca4409558e27dcdf8b4793fa47b964f8183120fbbb.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "aeb82d5f3e03fa18707a9b12128b6081dfa820ec22451c2c7e472d0e3f17b88b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "aa37907eb766f1a20ee9bb0d778f19907502a713516fac619837cec37c0fe4da", "generation_seconds_batch": 22.0146220800234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9062500665008855, "ce": 6.0146708488464355, "genuineness": 4.293961524963379, "identity": 0.7937932014465332, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aeb82d5f3e03fa18707a9b12128b6081dfa820ec22451c2c7e472d0e3f17b88b.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "af8d52ce7686b7eafab654f92616c3fde4656515cc35c372944e4a88a96ce086", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "17632a8aa4b7a725429c267fe61eb68050690832ec7327054e660002e706488c", "generation_seconds_batch": 22.272128722979687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.062326908111572, "genuineness": 4.068024158477783, "identity": 0.4298095405101776, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/af8d52ce7686b7eafab654f92616c3fde4656515cc35c372944e4a88a96ce086.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": "afb9e796071f03def104d7920ea624e1c60f666ba64c5cab3b7570fdba866eb7", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The road turns left behind the bridge.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "945b1939ee7482aef6676b383cf23b7baae1132fc547ea29e20b0f86aa2da77f", "generation_seconds_batch": 22.071688547963277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.750138282775879, "target_z": 1.4687501403907581, "ce": 6.507116317749023, "genuineness": 4.18726921081543, "identity": 0.2838215231895447, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/afb9e796071f03def104d7920ea624e1c60f666ba64c5cab3b7570fdba866eb7.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "aff176f7c8a69649c665c136c27a0f3defeaa6665f6240de6315fcc28c30eaee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5b1150d58e6672fe21ae4dddfdf8e15fea3c10a47bc6f2089bd8afa8b0c9bd62", "generation_seconds_batch": 16.549790724995546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.229123115539551, "genuineness": 4.369274139404297, "identity": 0.5596925020217896, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/aff176f7c8a69649c665c136c27a0f3defeaa6665f6240de6315fcc28c30eaee.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "b064bbf3d029b685aca8e22d6a631730830e76628af725e0fc6fff75519090a3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "502bb57e22c245e3fbc760acd4982e23e304fd86c74f9aba07afcbbf0c9bce48", "generation_seconds_batch": 21.411273667006753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.086154937744141, "genuineness": 4.325342178344727, "identity": 0.7152424454689026, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b064bbf3d029b685aca8e22d6a631730830e76628af725e0fc6fff75519090a3.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "b0b72e3ad04e717b27479683a980622ece6732fce658f31af46caac5245a9ad9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b059ed2c96612737995c37bbf288d2563a31364dd27ab782c971c9408010131e", "generation_seconds_batch": 24.349841281073168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.494641900062561, "target_z": 0.2236328422868549, "ce": 6.268836498260498, "genuineness": 4.425758361816406, "identity": 0.3277219235897064, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b0b72e3ad04e717b27479683a980622ece6732fce658f31af46caac5245a9ad9.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "b0f8ba50384ec079bad9b27bf130a48bab212ec4899dd8a3b44e9ab626991ceb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d32d71d5122c23b902e1c02faa78fab181083b98d9be39a765977cd4d8107077", "generation_seconds_batch": 21.063573995023035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.229123115539551, "genuineness": 4.005263805389404, "identity": 0.24473005533218384, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b0f8ba50384ec079bad9b27bf130a48bab212ec4899dd8a3b44e9ab626991ceb.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "b0fc0d2ec39ece309b8cad49979573478b30598bf8211353b9093216b44331f3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 129, "seconds_generated": 10.32, "mp3_sha256": "ac7503dd36f8b5dadccb1d072869f82e25081364bea5f772b74c1d54f862355e", "generation_seconds_batch": 23.674761812086217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9687500221669618, "ce": 6.356205940246582, "genuineness": 4.331617832183838, "identity": 0.7079859375953674, "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/vn_S_FORM_reg/b0fc0d2ec39ece309b8cad49979573478b30598bf8211353b9093216b44331f3.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "b158b2c19196afb87774faa360bbb0336051d3dc400a04980b395676c96e62f7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a91c88b74db17d7c73251bb3f1d036c8125533ff0997b3af3c8d8410a6552df0", "generation_seconds_batch": 24.283673704019748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.104631185531616, "target_z": 1.820312479680285, "ce": 6.1020402908325195, "genuineness": 4.118232727050781, "identity": 0.8045878410339355, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b158b2c19196afb87774faa360bbb0336051d3dc400a04980b395676c96e62f7.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "b17ca604f0e62e3ad1bcc4d61bdace1beb16c834f7f843a7ee2e16fa21ba232a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "284e33b9be1191ac10732c08d04bb5e7a97e754894993b24bb8205110dec1b65", "generation_seconds_batch": 21.638561701984145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.222337245941162, "target_z": 0.9453123910124378, "ce": 5.95112943649292, "genuineness": 4.13078498840332, "identity": 0.48652589321136475, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b17ca604f0e62e3ad1bcc4d61bdace1beb16c834f7f843a7ee2e16fa21ba232a.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "b17f300dd12605e2138dfa313b4fd207e2d30e75c5032bef0e97f198e3c25d9a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6870fa9c2a6a8ee84e1eca65b28f767ee92b25ec4d98e27e31642979fe43a9b7", "generation_seconds_batch": 22.71272316505201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.078212261199951, "genuineness": 3.8420867919921875, "identity": 0.34248173236846924, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b17f300dd12605e2138dfa313b4fd207e2d30e75c5032bef0e97f198e3c25d9a.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "b1a5c21043436edad36c018affe6c5cf477a3bbb8304b199e35338cac9b9d4ab", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ffb39977a0c6d4501b3ed6b3ab511f47277861c4e517fec14eb5a25b390d11a8", "generation_seconds_batch": 21.968828371027485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.356205940246582, "genuineness": 4.469690799713135, "identity": 0.6535836458206177, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b1a5c21043436edad36c018affe6c5cf477a3bbb8304b199e35338cac9b9d4ab.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": "b3cd917860d159291f13dfb19508430175b0fb01e0ce8693d3c845b48e7a886d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "0cdc1d2b5491b0bbc6314fe5d56fc58907f4f9485750b256707ad003c86a3a54", "generation_seconds_batch": 20.55520050600171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.348263263702393, "genuineness": 4.193545341491699, "identity": 0.6045002937316895, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b3cd917860d159291f13dfb19508430175b0fb01e0ce8693d3c845b48e7a886d.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "b3ec7dceb93b14ae196eda23ee789576c28340d7e4818f52732b7e27506cc1ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "23aa96af8e5a27f94b8c2c25a4dc3cf016bbe7db7a12b5d8de5b33ad55988188", "generation_seconds_batch": 21.1845044180518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.252951145172119, "genuineness": 4.231201171875, "identity": 0.8100946545600891, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b3ec7dceb93b14ae196eda23ee789576c28340d7e4818f52732b7e27506cc1ba.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "498aa8d37d65acd1e8320ddaafd332d012d92c7607f70aa3b8d54f93fe7061bd", "generation_seconds_batch": 16.80658993893303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0608460903167725, "target_z": 0.7851562693960916, "ce": 6.316492557525635, "genuineness": 4.582659721374512, "identity": 0.5064005851745605, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "b43abe51327729bd260614a3cf74ad9d61c3d567e252ec94d07c8cd040e64613", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "35bd8c299f549a8731e73c58cbb180769535b9d9818853683bf6fa3f01b5bfa8", "generation_seconds_batch": 23.71613089600578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.7167618274688721, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b43abe51327729bd260614a3cf74ad9d61c3d567e252ec94d07c8cd040e64613.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "b55902ed8bdc50ecec1b555fdfe9146c452a789909236f5fb8fd45715c0a6d6e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "05b6e56d204da2e0f191e37e45542acee3a2a344958d1b2e35516212c7c4e2dd", "generation_seconds_batch": 21.90544672694523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.5556851625442505, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b55902ed8bdc50ecec1b555fdfe9146c452a789909236f5fb8fd45715c0a6d6e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "b58aeb642a38d1e5f67817459335531a57db28639a437954820dd4e7ae1d7b7a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bf427d43aa79b4b26eb8c34f824fb045ff249d8b85145356a2f6393cb3d85a02", "generation_seconds_batch": 21.7470355440164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.679239511489868, "target_z": 1.3984374833747786, "ce": 5.903473377227783, "genuineness": 4.281409740447998, "identity": 0.7231864333152771, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b58aeb642a38d1e5f67817459335531a57db28639a437954820dd4e7ae1d7b7a.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "b5e467731dcf233f729d544f48f6052354724663bb934f48e371e5e8a9b21da7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b854f269da18d301d58a24c1310c585fb4053b8d4ca552b3af6f58018c52bc94", "generation_seconds_batch": 16.787156021106057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 6.268836498260498, "genuineness": 4.331617832183838, "identity": 0.5779474377632141, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b5e467731dcf233f729d544f48f6052354724663bb934f48e371e5e8a9b21da7.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "b5ef4672314935a7a005b4ac72aacf9e76ffcaac0be09387ef28384275fe46da", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ef50252991e11e6c9035c2ea842606c1c2421947a0210498ecb65a3438daebea", "generation_seconds_batch": 21.674366755993105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466543674468994, "target_z": 1.1874999852220254, "ce": 6.348263263702393, "genuineness": 4.18099308013916, "identity": 0.29169732332229614, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b5ef4672314935a7a005b4ac72aacf9e76ffcaac0be09387ef28384275fe46da.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "b64fb78c674911b62aa8ebaa258e3e0840450d8c40e43f314d2f700b44ec6c89", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "07ed8da5db88ad3a78f272b93dc6fcf99f8676499fd8aa0076c0101014c70c38", "generation_seconds_batch": 21.607595283072442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.6332390308380127, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b64fb78c674911b62aa8ebaa258e3e0840450d8c40e43f314d2f700b44ec6c89.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "b6a96f6585e8b43b712988d270775369c233cd9d38b4d10c1e5352ea567574d3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6296139552f622939a7536e10e3c805a98eaf15db973cdcb692f0e6d4e9e7a7e", "generation_seconds_batch": 21.37317533302121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7890679836273193, "target_z": 0.5156249298046209, "ce": 6.125868320465088, "genuineness": 4.7332844734191895, "identity": 0.7759386897087097, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b6a96f6585e8b43b712988d270775369c233cd9d38b4d10c1e5352ea567574d3.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "b6afa3c9f2b346ba979d97da9811cab4a8b8bd2764cf5b568aee8ceb6c46c9cc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4beb5f56f20f2eec1e786c0ef7e2ca13c8d463c9205990ec632d88e5ca213f9e", "generation_seconds_batch": 16.605777362012304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.0305562019348145, "genuineness": 4.143336772918701, "identity": 0.687242329120636, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b6afa3c9f2b346ba979d97da9811cab4a8b8bd2764cf5b568aee8ceb6c46c9cc.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "b7d279c9673134f5dea7c598aa9bb1f0f699995d9539f9eab7078959aa07fa6c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c334c9ed1f2218f0d657ca8ea563a26f881c6dbe488b6f43076454e8af0a0964", "generation_seconds_batch": 15.653577949036844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 6.308549880981445, "genuineness": 4.381826400756836, "identity": 0.43735912442207336, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b7d279c9673134f5dea7c598aa9bb1f0f699995d9539f9eab7078959aa07fa6c.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": "b809105a8c244c3a7cca779899b51816deeec37d702d154b05de172a9cfecdbd", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42f942f218119e7a19fda0a8d50e14dbbdf7f20106766a1edefee665505bd46e", "generation_seconds_batch": 13.668674846063368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.190826892852783, "target_z": 0.9140625314031958, "ce": 6.260893821716309, "genuineness": 4.595211505889893, "identity": 0.6147477030754089, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b809105a8c244c3a7cca779899b51816deeec37d702d154b05de172a9cfecdbd.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": "b920c0f44e2d30af6944e1bb7296a52714d510c80307fe44396e755a34d8bcc1", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5188cf961d009824fcf27acac33952e6e3102c5d817301ae03fbceb2cbc32cf", "generation_seconds_batch": 16.0540003470378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.7559788227081299, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces. Y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b920c0f44e2d30af6944e1bb7296a52714d510c80307fe44396e755a34d8bcc1.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "b990ee0ee714c4772850f19704e3d66e1862cc45dba78843d2ea52752e5c5860", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "76783c86e624a0ebb81066133e58550beaf8aeb13d382a0db5e664ab4a482135", "generation_seconds_batch": 22.849052945966832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.2767791748046875, "genuineness": 4.432034492492676, "identity": 0.4705139398574829, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b990ee0ee714c4772850f19704e3d66e1862cc45dba78843d2ea52752e5c5860.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "b9a83239ea0d6802d3423eb1695af14dff47b65809c24eb8636431b978399c2a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d8d3415a31ff1ad490a4dc0fae21814ae4ce1ee183df92bfe96a51f933267c9c", "generation_seconds_batch": 22.67743274802342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.070269584655762, "genuineness": 4.093128204345703, "identity": 0.672088623046875, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/b9a83239ea0d6802d3423eb1695af14dff47b65809c24eb8636431b978399c2a.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": "ba0d0e11ac13aef79c3180e1d1c0d8c3dcb7a475cb61e8f943ccdf2e0b6280ad", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09ae55c68c543d3d5403e0415a3cff57570d0df84c01f5dbfa3030a1b7b803ed", "generation_seconds_batch": 18.731670623994432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.38797664642334, "genuineness": 4.268857479095459, "identity": 0.6488528251647949, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ba0d0e11ac13aef79c3180e1d1c0d8c3dcb7a475cb61e8f943ccdf2e0b6280ad.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "ba634af24cfd6d11405e33a1aa98eb3cf72f127a0fec2d386df05343d30b840c", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "5a3ed8e3f1148f767f60824bbeaebd4907eb88fc139194e661df8bc8c5a09911", "generation_seconds_batch": 16.605416303034872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6634843349456787, "target_z": 1.3828125535701576, "ce": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.6049656867980957, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ba634af24cfd6d11405e33a1aa98eb3cf72f127a0fec2d386df05343d30b840c.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e49fec0d8adf437e8a22b52c8d5a64977f92b8ba1a6ad2adcb62c154b313d14e", "generation_seconds_batch": 15.541187877999619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4609374390408552, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.4848131239414215, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "ba8bcb57ae84f7bfc116e04e6f34eaaefa7f1902cee79d0c2f662c19b27f9370", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3e397f9c16e0ef46a2fdc444e0a406bfc2d62894fb6530f12bf4008f89069156", "generation_seconds_batch": 17.23408327891957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0017638206481934, "target_z": 0.7265624279573741, "ce": 6.078212261199951, "genuineness": 4.356721878051758, "identity": 0.7717654705047607, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ba8bcb57ae84f7bfc116e04e6f34eaaefa7f1902cee79d0c2f662c19b27f9370.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "bb9d82b8db5a5addbea8198f2a68076913fd7ec9054c725611b2bd4106970741", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9c94fcfe64530dca4adca5e36d9b73726ab69f856f7aa209e89a8db1ae66661f", "generation_seconds_batch": 20.9997038890142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8205785751342773, "target_z": 0.5468750258614554, "ce": 6.427689552307129, "genuineness": 4.632867813110352, "identity": 0.21371828019618988, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bb9d82b8db5a5addbea8198f2a68076913fd7ec9054c725611b2bd4106970741.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "bbced735b82e92b4a8c572072ec45145ce236a46105c863b7f0d49320c6fd5ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "7b16c46b43df2b42c3e2bb5d1fe1f1b3a2357ff1b4b231e870d84e816b7437d8", "generation_seconds_batch": 20.62663569196593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.899812936782837, "target_z": 1.6171874464298424, "ce": 6.221180438995361, "genuineness": 3.8546390533447266, "identity": 0.26331207156181335, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bbced735b82e92b4a8c572072ec45145ce236a46105c863b7f0d49320c6fd5ba.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "bcb552c08fa7652c5bdf553c7a1e541e82bdd95221b5e5681fd8bb812120b4f5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7c7181817038372e107c0fa06c08fca81e165015f7c413ab99dea565b100b9e3", "generation_seconds_batch": 21.208012967952527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1834073066711426, "target_z": 1.8984373651509825, "ce": 6.038498878479004, "genuineness": 3.948779582977295, "identity": 0.6337392926216125, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bcb552c08fa7652c5bdf553c7a1e541e82bdd95221b5e5681fd8bb812120b4f5.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "bd4eb7adfe74667fedd0d4deedd31e3e941ae93240f96e9971b72a1055116422", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c045892326b0f0d833c61ded20b8e8f571aecac73e4064ebe09fd48311729263", "generation_seconds_batch": 21.44179926905781, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1750717163085938, "target_z": 0.8984376015985749, "ce": 6.260893821716309, "genuineness": 4.344170093536377, "identity": 0.29807889461517334, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bd4eb7adfe74667fedd0d4deedd31e3e941ae93240f96e9971b72a1055116422.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "bd6eaf9100e9692bde94f01d56c48e80ca204d3dd71607f0a1ef50b8c51e01b1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f717f1e43af5eada9a82a46bdbfd0ef6802600e4b3830f177418334bdfc1f780", "generation_seconds_batch": 23.414000008953735, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.175529956817627, "target_z": 1.8906251366962643, "ce": 5.990842819213867, "genuineness": 3.948779582977295, "identity": 0.7398763298988342, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bd6eaf9100e9692bde94f01d56c48e80ca204d3dd71607f0a1ef50b8c51e01b1.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "bd710927abe42f0275b5bb92ee44f6ba61df18bfef0932d1342563588c3d981c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5cf1504ee7029614b6252dc244331bd8c5fac447a4001eeee736c9c65482f17f", "generation_seconds_batch": 20.8394880509004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5847082138061523, "target_z": 1.3046876680994604, "ce": 6.284721851348877, "genuineness": 4.375550270080566, "identity": 0.1630782186985016, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bd710927abe42f0275b5bb92ee44f6ba61df18bfef0932d1342563588c3d981c.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "bdb0de867326503e22845d87d503694ee4709ec16bfa676b72bcb736b14e0ff3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f37dbd913ddad64173fff8cc8abc222b759949aef10b8e8af9846dee2b1df6fa", "generation_seconds_batch": 21.766492787050083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.062326908111572, "genuineness": 4.042920112609863, "identity": 0.5333109498023987, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bdb0de867326503e22845d87d503694ee4709ec16bfa676b72bcb736b14e0ff3.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "bece2b4f5de608b2f8a1b028a942691daa0a370f2207f29f5a481e7c2e0382c5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2e7909d122336396f1bccc413609841c600cd0f68a48a76e00e8c0837f667d80", "generation_seconds_batch": 19.311757037998177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.553128719329834, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/bece2b4f5de608b2f8a1b028a942691daa0a370f2207f29f5a481e7c2e0382c5.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "c007891aec34ea78332397af04773b0dd5c4f2297bf702028ac13b482395571b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "286e2633b5da3d467c709bc88741a798b54cedd1540712453ae95ab300dd9b08", "generation_seconds_batch": 23.224474410060793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5851664543151855, "target_z": 2.2968752031971498, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.5200731158256531, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c007891aec34ea78332397af04773b0dd5c4f2297bf702028ac13b482395571b.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": "c02a125ac6ea9e1e1c05f1a196734ff7fcfd6c55bf80edce9a6d3d552e045ab2", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0faf0ab844d273ff95f455113c75510d77646bcb005ce7261b3e76c7d5a3687", "generation_seconds_batch": 22.26443234004546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.38003396987915, "genuineness": 4.18099308013916, "identity": 0.5869075059890747, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c02a125ac6ea9e1e1c05f1a196734ff7fcfd6c55bf80edce9a6d3d552e045ab2.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "c03e45af18a31aa72c49e04a763f0cf5a4048a9a92d66921a48425b62bc261bf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b359e14e61c3134f8d78fcf728ee713a691b315dcb6e238529b5fa18e3de65b1", "generation_seconds_batch": 16.91690432606265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.828456163406372, "target_z": 0.5546874907637659, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.5251676440238953, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c03e45af18a31aa72c49e04a763f0cf5a4048a9a92d66921a48425b62bc261bf.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": "c043091cf3db20e6b8c4858e096a19a7d4d11e4dace03918e99e3d519f645a09", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cff8da068a61660ec472dd25cedfce45404a16fc519966f148c04bb7c9160a0", "generation_seconds_batch": 18.326804143027402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.459460258483887, "genuineness": 3.9676074981689453, "identity": 0.28040510416030884, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c043091cf3db20e6b8c4858e096a19a7d4d11e4dace03918e99e3d519f645a09.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "c04fdf8e1969c9d3f3013064e2fced8fd9f4c700ea3744df9c4bf680f8e12b0d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "73af4785335d143933ce9c0ed51f5856bec59b396672401a42efc7f7cebd47d6", "generation_seconds_batch": 28.09001506504137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.226276159286499, "target_z": 0.9492187416873893, "ce": 6.24500846862793, "genuineness": 4.469690799713135, "identity": 0.6307633519172668, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c04fdf8e1969c9d3f3013064e2fced8fd9f4c700ea3744df9c4bf680f8e12b0d.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "c05b7bba85c1480cba566aea86f9791df3ebf91a76e954cec2bc2c4ce173ccc3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "547cdad5df09e04b9bddfac2e361ab8377510f7e9b6c3641f789cee9702647b1", "generation_seconds_batch": 21.319448751979508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0214579105377197, "target_z": 0.7460937084369467, "ce": 6.324435234069824, "genuineness": 4.394378662109375, "identity": 0.1363636553287506, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c05b7bba85c1480cba566aea86f9791df3ebf91a76e954cec2bc2c4ce173ccc3.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": "c07238f8d1f2e1c16af3827404744a80d7e5867d2783bfff88926bb1dffb82d3", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Your message arrived after I had left.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f07fbf41db3e71e6990e88478a1783f6ad24641f9e5e11d4e0ba4f71d440b90d", "generation_seconds_batch": 16.0294995669974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0888760089874268, "target_z": 1.804687549875664, "ce": 6.41180419921875, "genuineness": 3.8169827461242676, "identity": 0.3822084069252014, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c07238f8d1f2e1c16af3827404744a80d7e5867d2783bfff88926bb1dffb82d3.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "c0b000f961a18ea8b09a524cac3e5bd49da8b02ee7035f57b4e188a291c97205", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2d04a4fc19a97825c745693f1184a6f25c9726e6868f05691fc42e811d4bac80", "generation_seconds_batch": 24.51028929289896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7140480279922485, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Pudemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c0b000f961a18ea8b09a524cac3e5bd49da8b02ee7035f57b4e188a291c97205.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "c1240bbc9deeb7d3b2f69fdef7b25b7c30cafa18732f1369d7826925b43eaea1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d1b256a5e12299982b8e279a4d07b8a27d6262dea7bb4e95f473a04aa051f74b", "generation_seconds_batch": 20.3644430070417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.300607204437256, "genuineness": 3.9927117824554443, "identity": 0.1559317409992218, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c1240bbc9deeb7d3b2f69fdef7b25b7c30cafa18732f1369d7826925b43eaea1.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "c13830223721e9aeb42b078f3f6689f601ad16078958a8893ee7e3ce4bed2aca", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a74bafcd68e1a429bd30e3661681e96ae63f9d5f6ae06f8eb0ab5cbe7deeee10", "generation_seconds_batch": 22.009892446920276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.213237762451172, "genuineness": 4.293961524963379, "identity": 0.612541675567627, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c13830223721e9aeb42b078f3f6689f601ad16078958a8893ee7e3ce4bed2aca.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "c17b2f2ded190fbdbd933035265754ca36aba92c48a3bc6ad8bb35fce9113c8e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b51e565e9721d0e7f3dd0c002458d5030e89a6cabd73a756a194341d06caa47", "generation_seconds_batch": 20.31884474004619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 5.927301406860352, "genuineness": 4.1998209953308105, "identity": 0.6075292229652405, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c17b2f2ded190fbdbd933035265754ca36aba92c48a3bc6ad8bb35fce9113c8e.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": "c1aaa1ae179939abd0e5b2dcfda6e275ee52074f40060446ebf0fd21ca5331b8", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ff499c3fa89c1695ff5adb719ba88f8478f3f5bf12da14a2059cfbca19d3538", "generation_seconds_batch": 22.26443234004546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.38003396987915, "genuineness": 4.243753433227539, "identity": 0.6146484017372131, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c1aaa1ae179939abd0e5b2dcfda6e275ee52074f40060446ebf0fd21ca5331b8.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "c27de82b666f0dfa59486165fcdd5b146ce2828d1dd9fe5c8fc71b2e47f1e276", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a4e14f8be0b3e02bac9821d19b75b4531b5aafad85b01b3c97fa1544744a59ed", "generation_seconds_batch": 23.532899847021326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.348263263702393, "genuineness": 4.3504462242126465, "identity": 0.6229037046432495, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c27de82b666f0dfa59486165fcdd5b146ce2828d1dd9fe5c8fc71b2e47f1e276.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "c27f52980da3edb52f568ab22cf6c9fe37ef8ad08d666ededfaff710791e71bb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e6fc66a6c0d095a2a752963ca81b71dc447f437f99b6efb07e8beb86ce7f68a4", "generation_seconds_batch": 16.91973064094782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.3252290189266205, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c27f52980da3edb52f568ab22cf6c9fe37ef8ad08d666ededfaff710791e71bb.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "c31d4f7497d0fb82427f8574185109d5cbadf9086da77f843037cdeba0f4b7ed", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "075023f0069c74bc87b4908fede7b54cbb2f55ae6ddc6bbbb2b8837c0449a065", "generation_seconds_batch": 21.618538154056296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.260893821716309, "genuineness": 4.042920112609863, "identity": 0.3299124538898468, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c31d4f7497d0fb82427f8574185109d5cbadf9086da77f843037cdeba0f4b7ed.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "c32082e7e678ee82414bb977aa1c87d26981b6bff3cb20ed1a0296cf721b0b4d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "253a7f0f9b6c2e30be8723bb050a6f7b97d8d532676b52f1f1ded541d8581686", "generation_seconds_batch": 25.605192051036283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.086154937744141, "genuineness": 4.193545341491699, "identity": 0.7891867160797119, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c32082e7e678ee82414bb977aa1c87d26981b6bff3cb20ed1a0296cf721b0b4d.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "c327dee8ee016d7a77c13c795bcfebee49127776815aedeaae506bd8668560ab", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4c5d8e84815714dcd32f3a69f0ff1f297be7b24da5c87f1832a6d0d914feb149", "generation_seconds_batch": 22.638845414971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0844788551330566, "target_z": 0.808593664103023, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.7500993013381958, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos.", "wer": 0.23529411764705882, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c327dee8ee016d7a77c13c795bcfebee49127776815aedeaae506bd8668560ab.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "c36cab6be1a34780e2ba10283ed9d3faf6a5d8a3a4b0fd2d899b30f73fa89a2e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dd95bfe4b26e6c7f4a6408f4416dbfa42fa6c949efbc7fc3a24b2851b0d6ca91", "generation_seconds_batch": 24.828868657001294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.230215072631836, "target_z": 0.9531250923623408, "ce": 6.133810997009277, "genuineness": 4.193545341491699, "identity": 0.785060465335846, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c36cab6be1a34780e2ba10283ed9d3faf6a5d8a3a4b0fd2d899b30f73fa89a2e.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "c373482584673892e9c5b593bdbacb11dd35c3f4f24b4019ca3edd67130be1a5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9b5982901db7881e7f28fac0fd1a7f12e958a428b3f48074b5f75200650b0883", "generation_seconds_batch": 22.514748272951692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8289144039154053, "target_z": 1.5468750258614554, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.4276580214500427, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c373482584673892e9c5b593bdbacb11dd35c3f4f24b4019ca3edd67130be1a5.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d538b273ba683ded5e5f0bed6e1010bde1ac8519655a814c93add1e31923f2b", "generation_seconds_batch": 15.745734230964445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 5.95112943649292, "genuineness": 4.281409740447998, "identity": 0.6219621896743774, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "c44ae122791d85ba35fa9a6ac7949bebb0825433e51a9111e884724ea51b822e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "19f2cc344f849876e5c8b0e3f5ed57d50539d81f00701f654195816073e3056f", "generation_seconds_batch": 21.44179926905781, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8520890474319458, "target_z": 0.5781250036944936, "ce": 6.4912309646606445, "genuineness": 4.494794845581055, "identity": 0.41157013177871704, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c44ae122791d85ba35fa9a6ac7949bebb0825433e51a9111e884724ea51b822e.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "c475af4881ff7dbcb26057ed03e6a3df8850923b1955fcf7b358ba040d5663be", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "aa3800d7852fc3c4e311d44a226fa538dc9cc396685b696b3b17f82131027ced", "generation_seconds_batch": 22.239636090002023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.157639026641846, "genuineness": 4.18099308013916, "identity": 0.8432633280754089, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c475af4881ff7dbcb26057ed03e6a3df8850923b1955fcf7b358ba040d5663be.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58f0bb48cfd7b34584000f152a7b57154290dc164ec6ad8658793eb9a19174e8", "generation_seconds_batch": 16.552686794893816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.7549945712089539, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "c4a8c8c1712be6f9b344c5e8933ef290027b38d6cacb077dd4e14c715c3b5e92", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8e06ac4c64db2e38e5cbc93026ddc0fddcb6e753b1739f7dd66ccfa514253c82", "generation_seconds_batch": 16.710920891026035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0022220611572266, "target_z": 1.7187499630550638, "ce": 6.173524379730225, "genuineness": 4.137060642242432, "identity": 0.5072330236434937, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c4a8c8c1712be6f9b344c5e8933ef290027b38d6cacb077dd4e14c715c3b5e92.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": "c4f8e6aefa32fdc01dd60fe96f5f8d82011867b04565141e9e53153b257f7227", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "352bfc0ba7e95d34d9ed0d4c3e2c23092d876383b4666cf0700dc092a46f4285", "generation_seconds_batch": 16.566208334057592, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.171132802963257, "target_z": 0.8945312509236234, "ce": 6.332377910614014, "genuineness": 4.432034492492676, "identity": 0.6438114643096924, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c4f8e6aefa32fdc01dd60fe96f5f8d82011867b04565141e9e53153b257f7227.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "c57423e1c4d00478466eea2fad02a8b2ad141e5a4c8c5a7cecbe13cd1c40136f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ff5e2f8fcbf71dcb8bb3d24b019d5d310b4dd6eb37ba9b37da4af817aeabfbfe", "generation_seconds_batch": 24.176812999066897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 5.780362129211426, "genuineness": 4.218649387359619, "identity": 0.5959711074829102, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c57423e1c4d00478466eea2fad02a8b2ad141e5a4c8c5a7cecbe13cd1c40136f.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "c59a70501cf2d81358f2d9f86f97879585c522168da75d1f58b9a40789d2b950", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "3ce913a366923f06b0854388933eed0f9aa4d8faa93d1ca95388aa7ec78f5cd9", "generation_seconds_batch": 20.202318134950474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 1.179687520319715, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.765457034111023, "hypothesis": "De deer staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c59a70501cf2d81358f2d9f86f97879585c522168da75d1f58b9a40789d2b950.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "c66222d5c312d50120755f2e7ed63fbfc1d3a18956731828bb60fe2360fdcd93", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "0d9db4e68da84a135869dbfe27668947375623754fab6bd61c756888d2e9e953", "generation_seconds_batch": 21.33407437708229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.2890625018472468, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.8078590035438538, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c66222d5c312d50120755f2e7ed63fbfc1d3a18956731828bb60fe2360fdcd93.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "c6bcfcc97f2cd471fa97fedc1ce4ea47915d53d7129603c05396e7b8ac8833a6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2e0537cf9784bdce7abcd1336d2175fe665939d50813d817c43cc55c19e891f1", "generation_seconds_batch": 21.549633068032563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.086154937744141, "genuineness": 4.6203155517578125, "identity": 0.7483037710189819, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contentait.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c6bcfcc97f2cd471fa97fedc1ce4ea47915d53d7129603c05396e7b8ac8833a6.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "c7759279d9d6f1fe98188d5b92cd1ce0eb2b4b95bceeb484a77c3b1565e9c03d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4ccf6e655a0c8db953a1e130a83facc0124015b4edabda135245f1ae182af35b", "generation_seconds_batch": 19.58858231897466, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3405017852783203, "target_z": 1.0625000738898727, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.6882392168045044, "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/vn_S_FORM_reg/c7759279d9d6f1fe98188d5b92cd1ce0eb2b4b95bceeb484a77c3b1565e9c03d.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "c791cfe1f6c39336b32da0ad7286cbb41a51de763141843ae11c69027f839d1b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f5681e9e73458793d31d609494788a45c23c374c91805522f679103566d14431", "generation_seconds_batch": 21.930678023025393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.156249889165191, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.6754485368728638, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c791cfe1f6c39336b32da0ad7286cbb41a51de763141843ae11c69027f839d1b.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "c7ea8711d289e092efe355beb3db974d0c8f5dd82005cc4af86d68e31207f832", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6fdb9e825671b0b0fcf11bbf238c266746f85c5f0e252958035dca9577ceffe4", "generation_seconds_batch": 15.412823204998858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 1.9296874612078168, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.6284219026565552, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c7ea8711d289e092efe355beb3db974d0c8f5dd82005cc4af86d68e31207f832.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "c801dce893ca7a31717b359a061bd504b7337ba82b246a57d1b156da9c16d627", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d11bbda0ef5c8d5ab031dbbc6f1b1388d7160be8a86d4273ecfb5abecd2999b0", "generation_seconds_batch": 23.690930479904637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.281249800497344, "ce": 6.3641486167907715, "genuineness": 4.243753433227539, "identity": 0.5897818207740784, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c801dce893ca7a31717b359a061bd504b7337ba82b246a57d1b156da9c16d627.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "c814da4fa2bb7ff95b468b272e282c6c2ef7db22f58049e620534e4250d36129", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "49f52e0e62c0007bb88031b49ab1986a2622000ad857e28668309c0c0da7537d", "generation_seconds_batch": 23.21977522096131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.284721851348877, "genuineness": 4.482243061065674, "identity": 0.6919445991516113, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c814da4fa2bb7ff95b468b272e282c6c2ef7db22f58049e620534e4250d36129.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7843a0d06cfde7c302a8af408b3febbe13276d8f0a5b26d204ec209ed37c0844", "generation_seconds_batch": 15.010244044009596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3139488697052002, "target_z": 0.04443358624555981, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.7356374263763428, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar, wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80fe2cebdee89530a5f2d8cf218e45eca65f48bd3c663ec2b6637583af1743be", "generation_seconds_batch": 16.32004315801896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.4515180587768555, "genuineness": 4.281409740447998, "identity": 0.5027689933776855, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c8dfbe03a5e7e633e9c4b3af79735da6d0b5616585ecbce8a0e67d48e56eef5e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a66218a256f126f209b955e34195efafb4ca7c13960804c16da72c26af4e5041", "generation_seconds_batch": 16.856845609028824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.230215072631836, "target_z": 0.9531250923623408, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.6071962118148804, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c8dfbe03a5e7e633e9c4b3af79735da6d0b5616585ecbce8a0e67d48e56eef5e.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "c8e0d36f85eb4010893c3b0c7874a40f15ebcaa66bc36af15427541462170b32", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "a08a4e31a9570343339e8d639c727f3437b2bcfeddc3c610f996723ee72ec51a", "generation_seconds_batch": 22.33321251499001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.1562501256127835, "ce": 6.117925643920898, "genuineness": 3.9801597595214844, "identity": 0.24084684252738953, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c8e0d36f85eb4010893c3b0c7874a40f15ebcaa66bc36af15427541462170b32.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": "c972c0f58acd36aade114364613a30f3e260c55deed74d83435a26fc56c8c954", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1aba5020dc0f93af8dffab033e68f324ed705ea81e72990276a84fce303592e", "generation_seconds_batch": 23.65389902004972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.1894097328186035, "genuineness": 4.457139015197754, "identity": 0.8697971105575562, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c972c0f58acd36aade114364613a30f3e260c55deed74d83435a26fc56c8c954.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": "c9a3fd73cf6f2fb7ceffa7e2c1def6769d197d7864c94089cd2f6ad43649d5bc", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b84917b4be1ec0080b00c45001bd9e87a36a910062de74e06b980ea5aeace56", "generation_seconds_batch": 23.705865920055658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8525471687316895, "target_z": 1.570312420568387, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.6878038048744202, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/c9a3fd73cf6f2fb7ceffa7e2c1def6769d197d7864c94089cd2f6ad43649d5bc.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": "ca1f4e2758e67c031d3bb12e0775da54e01459376480886ef15c6ad2908d268b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad1312bd6fe2e4de4eca633ef4ed3542874991ba55d4159b4de24a30c090dc43", "generation_seconds_batch": 14.971647138008848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.213237762451172, "genuineness": 4.494794845581055, "identity": 0.728746771812439, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ca1f4e2758e67c031d3bb12e0775da54e01459376480886ef15c6ad2908d268b.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "ca37092b6f0f122a53c7fcf1b8edb4be5409bff90f24519ab6a2864fdc5d1596", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "641505f1b6069ba944d6043ce96168cded4862d6593894b3ea5d0eb50bce0a04", "generation_seconds_batch": 22.316030805930495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1912851333618164, "target_z": 1.9062500665008855, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.7148256301879883, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ca37092b6f0f122a53c7fcf1b8edb4be5409bff90f24519ab6a2864fdc5d1596.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "ca51bc8e8f9d41a753cfd0ab9274b8777a576785a755c451915f6e98efd67823", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "fbf88d1bb07e5d1ee3050e3e764bb867d31340ee073b4e8bd4455e5011e8e744", "generation_seconds_batch": 21.53510856302455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.372091293334961, "genuineness": 4.545003414154053, "identity": 0.5324927568435669, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ca51bc8e8f9d41a753cfd0ab9274b8777a576785a755c451915f6e98efd67823.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "cabfe6d97e5a1565aa5e073d23a66baf25e8360c1859e2cb12708adc9729f303", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d75b827f3fa4f55874a14af8d6be6bb2905596ad458085c619445d5197615969", "generation_seconds_batch": 22.433946365956217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.25384783744812, "target_z": 0.9765624870692723, "ce": 6.09409761428833, "genuineness": 4.381826400756836, "identity": 0.7060688138008118, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cabfe6d97e5a1565aa5e073d23a66baf25e8360c1859e2cb12708adc9729f303.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "cacfae893560fca9a1d60154e03356ed98eb1cdfc551871677335005ee2f6304", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a41a84d4b5d966d7067016744354368a6c379e3bb4d6b405eae483fa320f31ce", "generation_seconds_batch": 21.159915798925795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.316492557525635, "genuineness": 4.419482707977295, "identity": 0.28068625926971436, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cacfae893560fca9a1d60154e03356ed98eb1cdfc551871677335005ee2f6304.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "cb7b92e21886b8f61a24a0db2814e5d1659bf164b64005e8d0bb1ba1f1e49a4d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5abf7b1b22beb39e51c67802d507e0c87f36ac99f9da7a0cab44feb3244024d1", "generation_seconds_batch": 19.311757037998177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 5.990842819213867, "genuineness": 3.9676074981689453, "identity": 0.673723042011261, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cb7b92e21886b8f61a24a0db2814e5d1659bf164b64005e8d0bb1ba1f1e49a4d.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "cc9ae4c8263d27698728329102a8cff5cc24124a68405f0dc372def5f7b2d6ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 130, "seconds_generated": 10.4, "mp3_sha256": "2c2894e996e9ae74e5e0f3cfe3406f2a93fd6b078b9933787242609db072351d", "generation_seconds_batch": 27.49297992908396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.054384231567383, "genuineness": 4.457139015197754, "identity": 0.6538716554641724, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cc9ae4c8263d27698728329102a8cff5cc24124a68405f0dc372def5f7b2d6ba.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": "ccf68f59537d9c6e8cf0a3513272c58dcbedbc81e08371e1128d898259dbed51", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f781631fc5c99797dab1fbce4f218d302c6301563a1d1df12b5fb8bf6a2255d", "generation_seconds_batch": 21.715713114943355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.221180438995361, "genuineness": 4.193545341491699, "identity": 0.7888700366020203, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ccf68f59537d9c6e8cf0a3513272c58dcbedbc81e08371e1128d898259dbed51.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "cd7479349813835310213a4321f40178dea8dfa0cf092bfad1edf1458c05c3c4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "59649693414ca727d3c4441b645be536ad0fc7fc36b9dc0cbb91fd8b7a034d3d", "generation_seconds_batch": 23.671440733014606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 6.09409761428833, "genuineness": 4.243753433227539, "identity": 0.7084925174713135, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cd7479349813835310213a4321f40178dea8dfa0cf092bfad1edf1458c05c3c4.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "cd9293e65144cc139e8e97a57799b806bf5985b5c60c671bd896f1b0ac45be86", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "207c4c4933dd0b94202d5feda53f581e635fd014734380be4fb03b4c819bf06d", "generation_seconds_batch": 23.16196779801976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 6.3641486167907715, "genuineness": 4.3127899169921875, "identity": 0.5633889436721802, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cd9293e65144cc139e8e97a57799b806bf5985b5c60c671bd896f1b0ac45be86.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "cd98a06b0acdd0f4aba25aed571c76407d4095324b99f574bf731e0238cbd96d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "758a98a0725a502aa28423ab1a851e96eb6ee5eb8bb560bafef54c8fc9270427", "generation_seconds_batch": 22.40363223804161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.167652130126953, "target_z": 1.8828124353463613, "ce": 5.847874641418457, "genuineness": 4.042920112609863, "identity": 0.7090825438499451, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cd98a06b0acdd0f4aba25aed571c76407d4095324b99f574bf731e0238cbd96d.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "cd9c232418a807b6af5d0d1d938071e9ea0a56ccb81dd2ca77138c013a9f2975", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "44a8525490a1075fd4a2f2a9df2447f579d9d6c4a6000d8ff07f49ec5f9eee2f", "generation_seconds_batch": 23.85829269804526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.054384231567383, "genuineness": 4.118232727050781, "identity": 0.8238927721977234, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cd9c232418a807b6af5d0d1d938071e9ea0a56ccb81dd2ca77138c013a9f2975.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "cdaed260160a1b4fa06c41525ce6d7f014a51af987ceddd16f138bc8c7ad35ed", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b47caaf91fc230d8f43651423c3e4351d1a096b45e122a124481ffd3c8c52c04", "generation_seconds_batch": 15.1014803359285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.165581703186035, "genuineness": 4.375550270080566, "identity": 0.6835455894470215, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cdaed260160a1b4fa06c41525ce6d7f014a51af987ceddd16f138bc8c7ad35ed.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "cdd48c002328090301c575ca7d4adf9beeea3de8d07c39149fcceda63f668912", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2d0e1db57169e692c7136794e56a7ef28c06f5a0f7e888709761133af7c85de2", "generation_seconds_batch": 21.86483442108147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.078212261199951, "genuineness": 4.406930446624756, "identity": 0.42785778641700745, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cdd48c002328090301c575ca7d4adf9beeea3de8d07c39149fcceda63f668912.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "ce37cc4f89b9a142102dae7b4036b548184a7f5311def0d958e4049e412277cb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "3115ca96430070e1fe53c49b1227d91e7b969bec9940d566d4b3aaec4f17f3bb", "generation_seconds_batch": 21.761320462916046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 5.824047088623047, "genuineness": 3.9299514293670654, "identity": 0.6646105051040649, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ce37cc4f89b9a142102dae7b4036b548184a7f5311def0d958e4049e412277cb.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": "ce3c97f3e0b2c5af729188ff693802f981a06b77c58d51086fd575902bc77e38", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bed07514eb831a2ceb1531e47b3b4e9f1a1f3980e4bbd0d7cbcdd43f5761c78", "generation_seconds_batch": 16.03667499497533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.173524379730225, "genuineness": 4.2374773025512695, "identity": 0.5163908004760742, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ce3c97f3e0b2c5af729188ff693802f981a06b77c58d51086fd575902bc77e38.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ce77b63300cac738db293fefe7b93b82c9ed072a3984138128d6d465761929b3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "4b513053cdb3cde0af571523e55914c0c0208ffd79b14c751838436f18ec5f44", "generation_seconds_batch": 21.387530538951978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.157639026641846, "genuineness": 4.6203155517578125, "identity": 0.7474231123924255, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ce77b63300cac738db293fefe7b93b82c9ed072a3984138128d6d465761929b3.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "cf2a1c279a9786ccdf3e668dfc0a8501fe69c2534ef37992e342229862fe9bf3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "21bd2eebaa0e391531b4d7fbd2667b91969283f6d238e4e843d844ec62738996", "generation_seconds_batch": 22.291195678990334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.52956485748291, "target_z": 1.2500001773356944, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.13424517214298248, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cf2a1c279a9786ccdf3e668dfc0a8501fe69c2534ef37992e342229862fe9bf3.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6aeefd6547e471ee9e7bd0186f244eb86197a0418b372ac2453aa82163378493", "generation_seconds_batch": 18.05133886402473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.23706579208374, "genuineness": 4.15588903427124, "identity": 0.5831289291381836, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "d03b5ff94c250111faeb5dcd1556b1d5f6d13dfefd53fb32de330418cd98283a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4d981dae05452ed66bf5769befdb58065518bcb0450ba18c176d9d934b8a0f56", "generation_seconds_batch": 21.633461110061035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 6.1894097328186035, "genuineness": 4.2751336097717285, "identity": 0.8177796602249146, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d03b5ff94c250111faeb5dcd1556b1d5f6d13dfefd53fb32de330418cd98283a.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "d05f8a4fd1db61f84f5bfcb5c3d4bf65258ea5dced993010f29fa961be94c22d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8ba25d710aa487c2531fea34d9b7324c39606281e139d2dc55d3f31488cc2a4c", "generation_seconds_batch": 21.54012837598566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 6.173524379730225, "genuineness": 4.545003414154053, "identity": 0.7008012533187866, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d05f8a4fd1db61f84f5bfcb5c3d4bf65258ea5dced993010f29fa961be94c22d.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "d0a0540fa42789e705c1b54793ce6ac36a5c24030540f99eb1fec71f7d0d0d50", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f93a2f888bf816c12ebf42bf6a001c318260ec6389daef2c31d02d197f3af02f", "generation_seconds_batch": 25.673564390977845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6398515701293945, "target_z": 1.3593751588632261, "ce": 6.09409761428833, "genuineness": 4.243753433227539, "identity": 0.7790313959121704, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d0a0540fa42789e705c1b54793ce6ac36a5c24030540f99eb1fec71f7d0d0d50.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": "d0ccdd4b4f17a9fb1e8ef6d0fa2ac5cc16c4a78dbd723cbe2f4e86824d7d50e7", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec88fedd8ff7a19e0d41c098be11f085968e2ccada2a04121b12ab2284d98e26", "generation_seconds_batch": 16.65113284403924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 5.879645347595215, "genuineness": 4.6203155517578125, "identity": 0.7850590944290161, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien, et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d0ccdd4b4f17a9fb1e8ef6d0fa2ac5cc16c4a78dbd723cbe2f4e86824d7d50e7.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": "d100f636b6c17a00b96d9e0cb94009b9caedc4ffc24e37f01ea0294009dda0e9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11cb72ebb8d63a74cb7995a215941f072eec9f14ad2ed2af5cc86a8d39c48d14", "generation_seconds_batch": 21.9024108099984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.625678300857544, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d100f636b6c17a00b96d9e0cb94009b9caedc4ffc24e37f01ea0294009dda0e9.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "d10ef9f51d078419a0c3ae8b842ac09b3f9565f1911d1180eb48614aebc872e6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4ac30bf33db854c3a8cc6f73e9dcc1fd0914ef153b4734a12dc5d499a0b1cc71", "generation_seconds_batch": 22.4542823350057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 5.831989288330078, "genuineness": 4.344170093536377, "identity": 0.8051372766494751, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d10ef9f51d078419a0c3ae8b842ac09b3f9565f1911d1180eb48614aebc872e6.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c815e4cd132c8b1ee0eb7345d540fabc4cfb6036a572788f68fd0b9017ca85df", "generation_seconds_batch": 16.430591524927877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.332377910614014, "genuineness": 3.9299514293670654, "identity": 0.26258447766304016, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.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": "d14aa9cf408276f4ef2ed27a427eba17817067e2485fc43885f27b56a55df814", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48d8d94db5b85b1ae67282614fe2e397d8aa195a1aadbd564af96a801d310bdf", "generation_seconds_batch": 15.783159850980155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 5.982900142669678, "genuineness": 4.3504462242126465, "identity": 0.7841044664382935, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d14aa9cf408276f4ef2ed27a427eba17817067e2485fc43885f27b56a55df814.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "d217da5f39b5752dfd275b0940b699a54407e975c89017881035dc589858d09a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "2f86a268bb9125a469b328ddcbd79b0ccbbf9f4cdb318eb073dcb8f102a33e7a", "generation_seconds_batch": 22.239636090002023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.038498878479004, "genuineness": 4.149612903594971, "identity": 0.5907943248748779, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d217da5f39b5752dfd275b0940b699a54407e975c89017881035dc589858d09a.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "d21edc823da38de1a5218d66aad38f591e47e553c0144a4ea54c8e99c379a161", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c7696bff14f8786e3d3928c35547c074adcd76aa8198dd62cfd673a0893b58fe", "generation_seconds_batch": 15.436162595986389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.213237762451172, "genuineness": 4.093128204345703, "identity": 0.24047793447971344, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d21edc823da38de1a5218d66aad38f591e47e553c0144a4ea54c8e99c379a161.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04301964c5532c694aaa0fe92e2942c69bce5e62ceab804cddac29b54bf0ed49", "generation_seconds_batch": 15.219675200991333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1750717163085938, "target_z": 0.8984376015985749, "ce": 6.173524379730225, "genuineness": 4.519899368286133, "identity": 0.7706024050712585, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "d26cb698e6a1470ccdae8745349a3fbe5bf5b83f8cff54d8692ebc8ff8248e12", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "81ab7ba4c8624497c19ff205f30d2cf55be33924c0a17af2be925cbc7660d7d2", "generation_seconds_batch": 21.981262775021605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.156249889165191, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.7374511957168579, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d26cb698e6a1470ccdae8745349a3fbe5bf5b83f8cff54d8692ebc8ff8248e12.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "d2729e8ada93e4f75cab3ca7dee5a546941ffa714c78ac32abfc3642e8e9927a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4d4a8b55e57a7492a5c827393d10676c0c2224fc2aa64981d404c4ef934ea18c", "generation_seconds_batch": 21.90557630197145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.6355626583099365, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d2729e8ada93e4f75cab3ca7dee5a546941ffa714c78ac32abfc3642e8e9927a.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d286318d01ce469cee039bd9bea815eafa1143e25120bcd03a601a70c0928935", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bf6cce178b7ba986a42b7e24db1da912469c2ef7a09d3542ae3fcfd6429666b3", "generation_seconds_batch": 22.40363223804161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.046441555023193, "genuineness": 3.9173994064331055, "identity": 0.662350058555603, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d286318d01ce469cee039bd9bea815eafa1143e25120bcd03a601a70c0928935.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "d286765ff5bd213f767f02407b3563f9a7f6ad1b7edf65527fbf283689612bea", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3135d721818a1575d79ca39866c798d900dc823a8ddadae56e027003627ef203", "generation_seconds_batch": 21.39784261805471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.986008644104004, "target_z": 0.7109374981527532, "ce": 6.38797664642334, "genuineness": 4.457139015197754, "identity": 0.2688390910625458, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d286765ff5bd213f767f02407b3563f9a7f6ad1b7edf65527fbf283689612bea.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "d28e8151e52a8db7e3c7466d0914ccf300cc9cae6669d160f398a291f964382e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2247245381369057686f4d2ffd1b50ec4fea7424cfdd6fbba6bc1eddd809157e", "generation_seconds_batch": 23.068096271017566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.3641486167907715, "genuineness": 3.9425034523010254, "identity": 0.4197779893875122, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d28e8151e52a8db7e3c7466d0914ccf300cc9cae6669d160f398a291f964382e.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "d364551f0ec5f2ed55ebd44dd99579946d051976629ef2db255fc905c2d644cc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1ec7b151a5f26b01d9a589d4da0ba17451a97dba7c1e84e845b8b109bb9121dd", "generation_seconds_batch": 21.319448751979508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0529685020446777, "target_z": 0.777343804493781, "ce": 6.38003396987915, "genuineness": 4.356721878051758, "identity": 0.47932758927345276, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d364551f0ec5f2ed55ebd44dd99579946d051976629ef2db255fc905c2d644cc.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d79f1e7916e24b12611bb9d33858083d4c8dac86c1e1a93e33fe49e067c5f021", "generation_seconds_batch": 13.833044898929074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.23706579208374, "genuineness": 4.607763767242432, "identity": 0.7684164047241211, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "d4737b8ff1f6fd22ac5eea05ac0c8db9930e8c4aa19b1906f373034ccc929e10", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "03659fa3b76b47ec020e652a15c64aa87a4b4613e2bfd0b5e32241f611f642e8", "generation_seconds_batch": 21.994545589084737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6796975135803223, "target_z": 2.390624782024876, "ce": 6.157639026641846, "genuineness": 4.080576419830322, "identity": 0.583825409412384, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in zingen.", "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/vn_S_FORM_reg/d4737b8ff1f6fd22ac5eea05ac0c8db9930e8c4aa19b1906f373034ccc929e10.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": "d4ef3172de15867a8c2cad24b5f348b1ba06c18adbe2da0e01ef96f6300f408c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c9807a83be9d513ddaf92e2b8f6644ad449943dc7a3d96539d9ab20bfd88636", "generation_seconds_batch": 22.597686166060157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.049487829208374, "target_z": 1.7656249889165192, "ce": 6.141753673553467, "genuineness": 4.143336772918701, "identity": 0.8580181002616882, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d4ef3172de15867a8c2cad24b5f348b1ba06c18adbe2da0e01ef96f6300f408c.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "d51cc47b8a464eea06a2f6eaa6f2412189008548818b17469e2a3db41f79152f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bcfd7dd24aef05acc425cbae180d3211a6c76ca6a00e879cf2dcf786a6866891", "generation_seconds_batch": 16.51503476803191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.0146708488464355, "genuineness": 4.017816066741943, "identity": 0.3513835072517395, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d51cc47b8a464eea06a2f6eaa6f2412189008548818b17469e2a3db41f79152f.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d532323e139d383c797cf760a891fa6ac31956bc124507dfc19f1acf8314adfc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8053251e27db0e57ecb5e18ecf3d24f9c8f8a935e32ef2f97f311200b9bdda1e", "generation_seconds_batch": 21.09964662406128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4276137351989746, "target_z": 2.14062495936057, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.5998920202255249, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d532323e139d383c797cf760a891fa6ac31956bc124507dfc19f1acf8314adfc.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "d5b0f62ca63f67e619d4470bbd8220d3f911a68489e38d2ee1f0a5db83915126", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6b5551f77b11bdc2e087709b511cf44ef56e9150625e9e066677b8a9a2c4f496", "generation_seconds_batch": 21.958985276985914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2341537475585938, "target_z": 0.9570312065896999, "ce": 6.062326908111572, "genuineness": 4.400654315948486, "identity": 0.49230918288230896, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d5b0f62ca63f67e619d4470bbd8220d3f911a68489e38d2ee1f0a5db83915126.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "d5d3a26947c160fdc961bec383891f71d6deccaf61ea59e5ba277210b6c9a435", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4187a4879031ffc540d014eb11940a9c65907b0c9d5ada471caf76a91929fddc", "generation_seconds_batch": 20.044202320976183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.0305562019348145, "genuineness": 4.394378662109375, "identity": 0.7876393795013428, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d5d3a26947c160fdc961bec383891f71d6deccaf61ea59e5ba277210b6c9a435.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "d6458d6ae72003df62312258545e855e2fd3ba6483a2e25fe10fdc4cb08f6f6c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0d40b934ea3ee959ac5d62aa7d0f60757d464db7fba6d7872b2c1cf70cf8d7ad", "generation_seconds_batch": 16.605777362012304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9470787048339844, "target_z": 1.6640624722912978, "ce": 6.1020402908325195, "genuineness": 4.306513786315918, "identity": 0.7208873629570007, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d6458d6ae72003df62312258545e855e2fd3ba6483a2e25fe10fdc4cb08f6f6c.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d650382830b76bfa115f76e268ad954cfa688dee1701d30dc66d0d3ea64bcd81", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1b69db9a7dac9eb98a0b9036a659ff9b5311a03a27bbd69f59af9a87be79481b", "generation_seconds_batch": 16.775594857987016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.379889965057373, "target_z": 1.1015626348490175, "ce": 5.784333229064941, "genuineness": 4.325342178344727, "identity": 0.7261877059936523, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d650382830b76bfa115f76e268ad954cfa688dee1701d30dc66d0d3ea64bcd81.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d6e98c9089fe61b9e54ad3e3f7142c0e9f49c7aa3f606b9387f741546b1a340a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4b3e81be650d33cc52696ff27ec693714b0a9482188a13daa0a654e4e3000aa5", "generation_seconds_batch": 21.960687801009044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 5.863759994506836, "genuineness": 3.8420867919921875, "identity": 0.6865007877349854, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d6e98c9089fe61b9e54ad3e3f7142c0e9f49c7aa3f606b9387f741546b1a340a.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "d6f7a45656043366eb73fbbdf2905c4747fbc36cf5fcc0a9d0ca39f69ff709a2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bcb8c6295b16fc0f4eb55137a29c560407cbb2a02a797265821e6dabaef6717a", "generation_seconds_batch": 21.497531311935745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 5.903473377227783, "genuineness": 4.143336772918701, "identity": 0.8000702261924744, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d6f7a45656043366eb73fbbdf2905c4747fbc36cf5fcc0a9d0ca39f69ff709a2.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d711050cb90ac9f027ebf974efbbecc9e22134f72753774cdabc02781010e20b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 130, "seconds_generated": 10.4, "mp3_sha256": "ef340ef9d245ab6445a60511ec21c3ccad7989a785f264266087260002b7b9d9", "generation_seconds_batch": 27.433922859025188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 5.8875885009765625, "genuineness": 4.174716949462891, "identity": 0.6352322697639465, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "wer": 0.06666666666666667, "duration_error_s": 0.20000000000000107, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d711050cb90ac9f027ebf974efbbecc9e22134f72753774cdabc02781010e20b.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "d76f0404c9187898169b357a193d2b9c9931b433e8637c36616f0da7847f2a0c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "70a7064d27cc4add64182d39b8d62e992dceefbe1cdf353324ba01dfdd12fea2", "generation_seconds_batch": 20.3730546340812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.260893821716309, "genuineness": 3.9173994064331055, "identity": 0.23694609105587006, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d76f0404c9187898169b357a193d2b9c9931b433e8637c36616f0da7847f2a0c.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "d772f1f43525259d4a23dc99a204107ce0a953f7d914348d856efc7587895bc6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a056fc93d2f6ce872ec78950b8ca4f58adc652d1b70f69acedb7c532a7b72556", "generation_seconds_batch": 21.98617203196045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.0146708488464355, "genuineness": 4.055472373962402, "identity": 0.49655723571777344, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d772f1f43525259d4a23dc99a204107ce0a953f7d914348d856efc7587895bc6.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "d7f0d5d2aa809ae55d59795e4e798fc37d3b5b48982f068beba73a0c3baabc2a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3c12b64488d04c87b957cc1baac5d04d7d6e776d5761b440e476976cdda9cfec", "generation_seconds_batch": 21.525089472997934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5453200340270996, "target_z": 1.2656251071403153, "ce": 5.855817794799805, "genuineness": 4.425758361816406, "identity": 0.8380278944969177, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d7f0d5d2aa809ae55d59795e4e798fc37d3b5b48982f068beba73a0c3baabc2a.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "d7fa02226a5b6c05b17031cec0d7948caf8ff41234d0f1150aa1d75ed73aee1c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6441dfbe72c96e0a8c8d6c541727192a0b02ff68bea5f65da7f16723413d1ae8", "generation_seconds_batch": 23.224474410060793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.268836498260498, "genuineness": 4.2374773025512695, "identity": 0.5896943211555481, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen erg samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d7fa02226a5b6c05b17031cec0d7948caf8ff41234d0f1150aa1d75ed73aee1c.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "d810ae782dab4d03547ae535d5d46502849f976acc2dda3baa75f8622e1ee50c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1f2ff3534c4749d9fe78f4159b001323e61704dbb3044de24080e9f8512a8dd9", "generation_seconds_batch": 16.179021560004912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6634843349456787, "target_z": 1.3828125535701576, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.690697431564331, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d810ae782dab4d03547ae535d5d46502849f976acc2dda3baa75f8622e1ee50c.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "d889c6a5e4ae4f94ce619c04e35df77b984d8c066333c935af3cb9d09eb6dcb9", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "99973fc0dd7720293f55e60ebfd36d09459e4562c5662a2102331f5a7ad063fa", "generation_seconds_batch": 21.163426148006693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710750102996826, "target_z": 1.429687579431613, "ce": 6.062326908111572, "genuineness": 4.344170093536377, "identity": 0.6202268600463867, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d889c6a5e4ae4f94ce619c04e35df77b984d8c066333c935af3cb9d09eb6dcb9.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "d8f9c3a67427fd5f7ad707bfbd54185ea77bac0393389712caf63803e7aa6fee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c8b16977b3c1c3090ce60dc360bfa20e831b8a9de794afce5a518c6b91b4cb12", "generation_seconds_batch": 26.06565226300154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.474879741668701, "target_z": 2.1875002216696178, "ce": 6.324435234069824, "genuineness": 4.080576419830322, "identity": 0.4939356744289398, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d8f9c3a67427fd5f7ad707bfbd54185ea77bac0393389712caf63803e7aa6fee.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "d9091e995236768f93c8cbca126bd20ac87c24c22bd270cd33569a93353aa7f2", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "89ddcb1f29b4c61d400bd843f99d06f8892a1702610405d7851a2fa3e82d5d5f", "generation_seconds_batch": 16.517090607085265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.252951145172119, "genuineness": 4.65797233581543, "identity": 0.7763767242431641, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle comptait.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d9091e995236768f93c8cbca126bd20ac87c24c22bd270cd33569a93353aa7f2.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "d95b2b5d268fbd0729fe777fb39471328fa396b6aa78ae92172de6b75aeeb6f6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16349fa7bb51da73d052fc739c872cc3c71ec9f645f193f5b902a904c4b4ac6c", "generation_seconds_batch": 23.083264735061675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.9999998817762038, "ce": 6.125868320465088, "genuineness": 4.6203155517578125, "identity": 0.8971105813980103, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d95b2b5d268fbd0729fe777fb39471328fa396b6aa78ae92172de6b75aeeb6f6.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "d971e0d42124dc051c84b13adf6f7996fabd7a84ae8ff98a42280bb28cd86e68", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "7c25159f4a552a4c14234f2fd0cad8bf6398b479a2eaa8004a3dd9a21c0654aa", "generation_seconds_batch": 21.07782974501606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.1894097328186035, "genuineness": 4.419482707977295, "identity": 0.48652341961860657, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d971e0d42124dc051c84b13adf6f7996fabd7a84ae8ff98a42280bb28cd86e68.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "d9fc3cdce8a9b07c7ffb0c86705b502eb72754f79d689c4089ea9c68ee335ecf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "32d34fb103a21e6dea495a2f957bde0e82d267fb7887e8d78105dadabf155c4e", "generation_seconds_batch": 22.350354934926145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.117925643920898, "genuineness": 4.20609712600708, "identity": 0.5239660739898682, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/d9fc3cdce8a9b07c7ffb0c86705b502eb72754f79d689c4089ea9c68ee335ecf.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "da7bf3580e3b894061170f9cf9c8d0f5205fc5a2257164cbf4c5329805c0f12b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "41cc41a47f890cb5b86555944d18b5e4c28dbf2fc0f78ca0bbf1d37ba5acbea1", "generation_seconds_batch": 22.18294645601418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.372091293334961, "genuineness": 4.1998209953308105, "identity": 0.5871701240539551, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/da7bf3580e3b894061170f9cf9c8d0f5205fc5a2257164cbf4c5329805c0f12b.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "da98eaab99200f09bbd68dd5194fbd1169a0e4d6ebf1001fd08411ed8a49fc50", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4ae9e94db21305c73de5ca9b91bf28fcb5fe79f81c282ce02a82e4f6ca5bbfbf", "generation_seconds_batch": 21.39784261805471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1435611248016357, "target_z": 0.8671875055417404, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.4677257239818573, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/da98eaab99200f09bbd68dd5194fbd1169a0e4d6ebf1001fd08411ed8a49fc50.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9f036fb1df4d709aa183041ee694be18b3d415ebdf0f341b45e99688185b4c6", "generation_seconds_batch": 16.725593436975032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2065820693969727, "target_z": 0.9296874612078169, "ce": 6.332377910614014, "genuineness": 4.086852550506592, "identity": 0.4560646414756775, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "dbfaf2000d3aa0e4c07fa7604fafe56abcdd3982d0bffb0f0bcf3072b6b98c69", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "08c616ae4e1a5a7f4c3d688acf2e5dc71d143edcd91fb052f240d6d6690549e2", "generation_seconds_batch": 15.789838057011366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.300607204437256, "genuineness": 4.18099308013916, "identity": 0.6225187182426453, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dbfaf2000d3aa0e4c07fa7604fafe56abcdd3982d0bffb0f0bcf3072b6b98c69.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "dcdc3cc868be0d2a0488bf65831aca888c7b9d29769e86d5c8e2fa3ad22d56d8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "06945b34772ad1e47ce398981fe534694191b8ec8eb7d69311417d740fc49cb5", "generation_seconds_batch": 22.19881454401184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2420313358306885, "target_z": 0.9648436714920103, "ce": 6.38797664642334, "genuineness": 4.362998008728027, "identity": 0.3341306746006012, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dcdc3cc868be0d2a0488bf65831aca888c7b9d29769e86d5c8e2fa3ad22d56d8.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "dd5edf4e8a769b1c645b226f36569414c8b163406fc0bb83ec2c7e114e0bc7d5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9d04e9ba8870cd28061d90e89b747c4cb13a76cab7d37303fb19b45f1477ef9d", "generation_seconds_batch": 15.773359431070276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.2343750110834808, "ce": 6.348263263702393, "genuineness": 4.3504462242126465, "identity": 0.7939668893814087, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dd5edf4e8a769b1c645b226f36569414c8b163406fc0bb83ec2c7e114e0bc7d5.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "dd6c4468fc043254dd584629898ee6bab100ec522ea16032897fe00c05ee1dfc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ce901685f750509a205864f9047ff8e9b80a1d6da90affe9b383046197da8730", "generation_seconds_batch": 21.638561701984145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3956451416015625, "target_z": 1.1171875646536387, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.7786573171615601, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dd6c4468fc043254dd584629898ee6bab100ec522ea16032897fe00c05ee1dfc.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "dd7be905b5046ea2575aaaf47adc36fd64c18a8b6a2983e4f87ff8d8bcda3272", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1f4252618132381c44c6ffecbc0067cdca7b55a6e3dbabff2321b8558d56a830", "generation_seconds_batch": 22.33321251499001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9687500221669618, "ce": 6.141753673553467, "genuineness": 3.8169827461242676, "identity": 0.30136698484420776, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dd7be905b5046ea2575aaaf47adc36fd64c18a8b6a2983e4f87ff8d8bcda3272.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "ddfb36ea4ed8c579e9f34532665fde7f8820cdcf96697d9ad2a6f7987fb45d14", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "d5bf330f5746f88178cdba740b3a0d7f8dd1196d08ef3024ded2e67e20d8dd33", "generation_seconds_batch": 22.79169623204507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.09409761428833, "genuineness": 4.111956596374512, "identity": 0.7736240029335022, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ddfb36ea4ed8c579e9f34532665fde7f8820cdcf96697d9ad2a6f7987fb45d14.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "ddfdbf088d94c7e6d74591ab591e50ca65b6cb7eef9f13b734d2663dedb92ea5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d062eacafec574d1d0e77e02b517e1629e924fe5ff1d4c4226213c3546f2c380", "generation_seconds_batch": 22.569629327044822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7974038124084473, "target_z": 1.515624929804621, "ce": 6.300607204437256, "genuineness": 4.124508857727051, "identity": 0.17370286583900452, "hypothesis": "I found the little box. It was beside the old window. We can look inside.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ddfdbf088d94c7e6d74591ab591e50ca65b6cb7eef9f13b734d2663dedb92ea5.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": "de8378fd7f201062e8d99cc98ab1bf14eea6008a4491567fb9c7e3585cb58f5a", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1c796b3c241fb41f4c7f9453fbaccc2d5db749ed15ddf689572ac37dc4302c3", "generation_seconds_batch": 18.731670623994432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.403861999511719, "genuineness": 4.250029563903809, "identity": 0.6113621592521667, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/de8378fd7f201062e8d99cc98ab1bf14eea6008a4491567fb9c7e3585cb58f5a.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "deed902330f6458bbf9b2472a8fd52d47fab99061a139da2454e505eec57ac65", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b34a1209237d5c760739be7d623374f33aa3f03ce2d81d144e3fc721b2b412f", "generation_seconds_batch": 15.79761787399184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 5.95112943649292, "genuineness": 3.998987913131714, "identity": 0.5937290191650391, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/deed902330f6458bbf9b2472a8fd52d47fab99061a139da2454e505eec57ac65.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "df3ac250c06ae4ac03cb1a819853a1670a10f11810663a1258c9fff97aa2892f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a6f4010bc0b1720b14db31f7935f3e87293d6365ee8d3221c0bfd50a9b6bd54", "generation_seconds_batch": 21.692288791993633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1435611248016357, "target_z": 0.8671875055417404, "ce": 6.086154937744141, "genuineness": 4.1998209953308105, "identity": 0.628265917301178, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/df3ac250c06ae4ac03cb1a819853a1670a10f11810663a1258c9fff97aa2892f.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "df6988e3962f354b6fb5c0f23a5268b2543b69d86fc51a5538ebeb252c8a6693", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ed6fc00df53d8e1e06f0a33967f1fcf4396f80b59dade6fb724aed41b63097ad", "generation_seconds_batch": 21.95830267400015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5531973838806152, "target_z": 1.2734373355950335, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.6827689409255981, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/df6988e3962f354b6fb5c0f23a5268b2543b69d86fc51a5538ebeb252c8a6693.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "dfaa68909c15b322e03a6a59cc804cf1de5112fb24d00ed6fa026e5230949b3a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5c7f1e57aff2e7cd0f7e35c49cac7aa708ef332500ff8a5946af689ea2c5025a", "generation_seconds_batch": 15.789838057011366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.316492557525635, "genuineness": 4.331617832183838, "identity": 0.8026273250579834, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/dfaa68909c15b322e03a6a59cc804cf1de5112fb24d00ed6fa026e5230949b3a.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "e014bdc69d523efc873597187e633d1a82ee5109955a98e02251851459eeb239", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d4730c20bd54ce64198f167d812485ddac57f7ab65aadb1e30324368e0698b97", "generation_seconds_batch": 20.8394880509004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.292664527893066, "genuineness": 4.124508857727051, "identity": 0.3864758014678955, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e014bdc69d523efc873597187e633d1a82ee5109955a98e02251851459eeb239.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "e04e38327fb2d4b15d890e13bf0551f05f3fc39d2d3d97e47c4eba83ec953914", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6fef38f94e9bc8db9d84fdb8ed54f87089c403fb2023236cd67a567d27067b1a", "generation_seconds_batch": 25.1623691880377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.165581703186035, "genuineness": 4.256305694580078, "identity": 0.6103940010070801, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e04e38327fb2d4b15d890e13bf0551f05f3fc39d2d3d97e47c4eba83ec953914.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "e076ddcb6813049135075658ce1f55e9317aa3d7b510918831bf9d1411b7dd3d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "10e139f456c0ac5c53b3631f25e404db70deb901dc508237323c6ffb2e74ed82", "generation_seconds_batch": 20.9997038890142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0332744121551514, "target_z": 0.7578125240142086, "ce": 6.483288764953613, "genuineness": 4.394378662109375, "identity": 0.36808353662490845, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e076ddcb6813049135075658ce1f55e9317aa3d7b510918831bf9d1411b7dd3d.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "e091cd2844d6528911b3afabd092ec8d3f36993e8a765e2546e1a777bbd24c56", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "5e7a6dcde83c7339a666741f5ce67c1d2f4076502fa52de809fadaff7fed5154", "generation_seconds_batch": 23.671440733014606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.205295085906982, "genuineness": 4.21237325668335, "identity": 0.6352468132972717, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e091cd2844d6528911b3afabd092ec8d3f36993e8a765e2546e1a777bbd24c56.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "e0e7dfa77c02b1b7e453179e5f884deffc5dfc381da9b17ddff64422eff30a42", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "75a3fafc152961b9290cc715841d3323e892df5ddacb55ab55016d39dceab12d", "generation_seconds_batch": 20.70256081793923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2149178981781006, "target_z": 1.9296874612078168, "ce": 5.919358730316162, "genuineness": 4.030367851257324, "identity": 0.6354720592498779, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e0e7dfa77c02b1b7e453179e5f884deffc5dfc381da9b17ddff64422eff30a42.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "e0e92f887aa7dab085560697f97ec44daea843a72e028a351b24f091d2beb38e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6b151f8e4b6e6496ed448ecd7e4c03456026bdb13ba6365ec3f480fb66954f65", "generation_seconds_batch": 21.39418767497409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.136141777038574, "target_z": 1.8515625757371195, "ce": 6.395919322967529, "genuineness": 4.3504462242126465, "identity": 0.5666683912277222, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e0e92f887aa7dab085560697f97ec44daea843a72e028a351b24f091d2beb38e.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": "e0fd8567cafbfe469bf8ad82d4184a5e4826bca146cf544306cb42b9ca258c55", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2e6fb7156ff6bfc75142f02e83d583eaf0d77a46ea14c29c314a3b1baeb3f3b", "generation_seconds_batch": 22.51586488704197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.9609375572646512, "ce": 6.403861999511719, "genuineness": 4.2751336097717285, "identity": 0.3129437267780304, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e0fd8567cafbfe469bf8ad82d4184a5e4826bca146cf544306cb42b9ca258c55.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": "e1373f1c8e80ee62c53bf764e696fdd43e7163e951f801f15f47299c3263ce60", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38101cbbfc3f1a8dc908e6c43af8981f887b26fc2b60d1506dc50ee157dfdbec", "generation_seconds_batch": 18.172833193908446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.459460258483887, "genuineness": 4.1998209953308105, "identity": 0.3658062219619751, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e1373f1c8e80ee62c53bf764e696fdd43e7163e951f801f15f47299c3263ce60.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "e1dc9606d0f3e5343a3be5db72018fb0f25f075b6312c1f7fe77041a2d60a452", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "cb213ffc2325fd32bfa87cf0e26de88f06f7bb4dc029a1b3f626d53da13f676b", "generation_seconds_batch": 25.17194919893518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3405017852783203, "target_z": 1.0625000738898727, "ce": 6.181467056274414, "genuineness": 4.457139015197754, "identity": 0.6247725486755371, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e1dc9606d0f3e5343a3be5db72018fb0f25f075b6312c1f7fe77041a2d60a452.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "e24495e174b06072f9a84f79196b59a58535a256c582e52066f699f06feb18ba", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "48bf40b2a533a6f710d6f230b30e4c1382cd6c4efb3f8defbc3674bcd44ed4cd", "generation_seconds_batch": 21.5460973510053, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.5780436396598816, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e24495e174b06072f9a84f79196b59a58535a256c582e52066f699f06feb18ba.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": "e261c04076adcff87e0b88da4bd0fdf0cc366da76bbed5e78c99a6274985590c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bfa058446ae715f821063dfb4ce751ad842af339105e9f3ed7966510964db413", "generation_seconds_batch": 21.482375354971737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411858558654785, "target_z": 2.1250000295559492, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.7583420276641846, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue, elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e261c04076adcff87e0b88da4bd0fdf0cc366da76bbed5e78c99a6274985590c.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "e2ad197220e0768f5cc9ec8fff459b14b83553922bc6a5935c3d7bf9af93e539", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "db9d8818854bacb087efb9a45275e672398f67f147426ddb2b4b507e1128913b", "generation_seconds_batch": 22.777148399036378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.348263263702393, "genuineness": 4.080576419830322, "identity": 0.26878148317337036, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e2ad197220e0768f5cc9ec8fff459b14b83553922bc6a5935c3d7bf9af93e539.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "e393831b7946688465d0dd14523e049a659f1e3efe38beb63e141ff994d32a09", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "301d521f8b2f74d486584a9535bdaf2cb93805f39252be10c2797eef1cc124ce", "generation_seconds_batch": 22.52458139101509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8210368156433105, "target_z": 1.5390625609591448, "ce": 6.348263263702393, "genuineness": 4.055472373962402, "identity": 0.2539413869380951, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e393831b7946688465d0dd14523e049a659f1e3efe38beb63e141ff994d32a09.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": "e3a8c4230034e8b03d69bc3817c61a7d3ac3cdea0fd855d1eaa84b45c9f8b73b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Escribí el número en el sobre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8566af9d96a342ed7183029a886224a0db2a9fba3509884bfd9fbd7f018c7a84", "generation_seconds_batch": 16.611532305018045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.308549880981445, "genuineness": 4.306513786315918, "identity": 0.7011082172393799, "hypothesis": "Escribí el número en el sobre. Lélelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e3a8c4230034e8b03d69bc3817c61a7d3ac3cdea0fd855d1eaa84b45c9f8b73b.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "e3e8120a29fe31faef127160d5bc06ed57e71b82476efb04d181d6bbdf9feac0", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "947e558af710931ab1966f814b9c58f70ab3470e16913015002cccf8bafc8a92", "generation_seconds_batch": 15.166684727999382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.270061492919922, "target_z": 1.9843751884191752, "ce": 6.419747352600098, "genuineness": 4.281409740447998, "identity": 0.5266513824462891, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e3e8120a29fe31faef127160d5bc06ed57e71b82476efb04d181d6bbdf9feac0.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "e3f9c4a4dc8391950e2970e22458eafa1ae547e0876ab0dce7483ec8d52f46d8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3a18161ed75108f744b35f6c997b40910e842221e71bc22ce2818964a42b4cd7", "generation_seconds_batch": 27.49297992908396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.181467056274414, "genuineness": 4.3504462242126465, "identity": 0.6000980138778687, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e3f9c4a4dc8391950e2970e22458eafa1ae547e0876ab0dce7483ec8d52f46d8.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "e41bc12939341ee00bc33074ee8127983aeaf0bcfe9719466bbf7ed668d065cf", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d00d9cb1d6040b82d32da9831e4d61a6900a3916be362e78a1a3bfbcb5bca273", "generation_seconds_batch": 20.870801943005063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.010558128356934, "target_z": 2.718750199502656, "ce": 6.062326908111572, "genuineness": 3.948779582977295, "identity": 0.5867724418640137, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e41bc12939341ee00bc33074ee8127983aeaf0bcfe9719466bbf7ed668d065cf.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": "e446a70fc7806d2e95f13171566efb632721fb7e44ccccdb49a216a04722de2b", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2e2e37de92ef37eeb67a65f014dbe0372d2c2300e902be5685dd4c8393e0b12", "generation_seconds_batch": 16.31743995693978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7658934593200684, "target_z": 1.484375070195379, "ce": 6.038498878479004, "genuineness": 4.532451152801514, "identity": 0.6882625818252563, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis 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/vn_S_FORM_reg/e446a70fc7806d2e95f13171566efb632721fb7e44ccccdb49a216a04722de2b.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "e4dde26a41282dca5c2a6b2066a1add2ff12fef0152df2540eec55801616a4ea", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "bf6881e229b8542d7da0a89aea2539066da99978adeb2a3edf37bdd78c0942b5", "generation_seconds_batch": 21.628867999999784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8087620735168457, "target_z": 0.5351562102841935, "ce": 6.117925643920898, "genuineness": 4.118232727050781, "identity": 0.8266881704330444, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e4dde26a41282dca5c2a6b2066a1add2ff12fef0152df2540eec55801616a4ea.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "e55de1cf6b864e8a6fb8335e77b210ba349b5f827a87e88b1bd33d0fc2746e32", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "836ea266e4669ea4a029260315d006cd4486e106e82908ef2e3a0e039e447f05", "generation_seconds_batch": 21.664702587993816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592585563659668, "target_z": 1.3124998965541783, "ce": 5.911416053771973, "genuineness": 4.306513786315918, "identity": 0.8338243961334229, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e55de1cf6b864e8a6fb8335e77b210ba349b5f827a87e88b1bd33d0fc2746e32.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": "e58f6a0760353ebb5ac43dd41bc535e9f1948a3bee68f524b6db474837602eab", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ce83cabc9552a7bb863b1839b936503f0cdf045a45d3ed59d47152b807fb0528", "generation_seconds_batch": 16.083552113035694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.23706579208374, "genuineness": 4.457139015197754, "identity": 0.8031939268112183, "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/vn_S_FORM_reg/e58f6a0760353ebb5ac43dd41bc535e9f1948a3bee68f524b6db474837602eab.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "e5a852ed074efc3a350578415229d236c017708996ca45e313632e8466e2ccc3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "d5b4f2eceb9a2d0a1d6692b100502d7b3f857dacb6dd1d5307378be25f96cc86", "generation_seconds_batch": 22.333251575939357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3089911937713623, "target_z": 1.0312499778330382, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.7601220011711121, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e5a852ed074efc3a350578415229d236c017708996ca45e313632e8466e2ccc3.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "e63bba69203adc48054d2f10f4013eabc808d30f3564473da31cf8ae6f948077", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c6d27347424b29bda4a751870291c6ee5b1008492c3d0621b75f65afc6dd9c24", "generation_seconds_batch": 13.600029004039243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5216870307922363, "target_z": 1.2421874759857914, "ce": 6.2767791748046875, "genuineness": 4.469690799713135, "identity": 0.8048692345619202, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e63bba69203adc48054d2f10f4013eabc808d30f3564473da31cf8ae6f948077.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "e69f6097786095c1b1c62ecb15ad4e2af8e50c407be71cdb878eb9606e57ddee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "53e533c6401ee4ccc7b5a54f856e9ee7cae738d40f59c5a9b14c159611d6db87", "generation_seconds_batch": 20.162616575951688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2774806022644043, "target_z": 0.9999998817762038, "ce": 6.308549880981445, "genuineness": 4.149612903594971, "identity": 0.3639652729034424, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e69f6097786095c1b1c62ecb15ad4e2af8e50c407be71cdb878eb9606e57ddee.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "e6d793fa884278a150fe52313227770bdac70817ce3a4650af128bacf428e743", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4bd6f9b7d7ccef1a2f66824393c71fe67530efd8b497ecc92d05fba857f1dd79", "generation_seconds_batch": 25.073003839934245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.6888585090637207, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e6d793fa884278a150fe52313227770bdac70817ce3a4650af128bacf428e743.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a02ab87727c7d030283a3644b3bb481181e4007058722f73af95b52f36d55f78", "generation_seconds_batch": 16.32004315801896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507884979248047, "target_z": 1.1718750554174044, "ce": 6.435632705688477, "genuineness": 4.20609712600708, "identity": 0.27393588423728943, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.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": "e780db2dbbf224186603aff4f16a40413d327d2122c27e9ef58d98a028536e1f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "105fca3525faad725d245eaafa047a53a5266bbc9d1033f25163dd8fb9886c86", "generation_seconds_batch": 23.622165640001185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.803318440914154, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e780db2dbbf224186603aff4f16a40413d327d2122c27e9ef58d98a028536e1f.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "e7f650d70455e8fd56ea0ee3e2b8e98d0799f0b22d8c6bb64f1cc1bf543cc25f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "38bdc21d3b6841625f7a0b80a31a4388f97551a6d403cf17bcf08952507080a7", "generation_seconds_batch": 15.436162595986389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.072662591934204, "target_z": 0.7968750849733536, "ce": 6.268836498260498, "genuineness": 4.362998008728027, "identity": 0.17259249091148376, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e7f650d70455e8fd56ea0ee3e2b8e98d0799f0b22d8c6bb64f1cc1bf543cc25f.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "e8b5ffbea57a9ab49b523e9d012f618500184af73d9d7846460edcc90e8145e6", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "80caf7fbb715e41282db838a2ac0b7c0fdcfe684134e25191120e65e98f9d130", "generation_seconds_batch": 16.079922795062885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.1020402908325195, "genuineness": 3.7918784618377686, "identity": 0.2758813202381134, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e8b5ffbea57a9ab49b523e9d012f618500184af73d9d7846460edcc90e8145e6.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "e90a44f5b879cdeab57c152fa48828ff741f10e3fb10ed30b63cd61c1df21496", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4ae50aecf9387ff2b3f05f00c24561c10437032a49cf68fa7f4e28b44a1784c7", "generation_seconds_batch": 22.265967768034898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2858166694641113, "target_z": 2.000000118223796, "ce": 6.356205940246582, "genuineness": 4.300237655639648, "identity": 0.5502162575721741, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e90a44f5b879cdeab57c152fa48828ff741f10e3fb10ed30b63cd61c1df21496.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "e9450b96844b8d11de6830510696499d41d07b0929c8dc41ae6ef7d7e406ebd7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b21111a49b94e8e739488af252e8173005131ebe10449df0d0ddf89bcc9417db", "generation_seconds_batch": 25.605192051036283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.561075210571289, "target_z": 1.2812500369449362, "ce": 6.157639026641846, "genuineness": 4.337893962860107, "identity": 0.8203585147857666, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e9450b96844b8d11de6830510696499d41d07b0929c8dc41ae6ef7d7e406ebd7.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "e97463efeedbe440c3cf21fb729aefa2fa05085fc9608ab7486ee24d8bebc4f8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9d30718b0bd3d490a4dbb75703adc72ba63c9fa2fbbe0285bc14f2b62157f77d", "generation_seconds_batch": 21.7227265490219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.569410800933838, "target_z": 2.281249800497344, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.5686905384063721, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e97463efeedbe440c3cf21fb729aefa2fa05085fc9608ab7486ee24d8bebc4f8.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd53e7a817813899bffc32a6aaac7e1d778dfdfd6cbd1354ed4edb82fa2e18af", "generation_seconds_batch": 18.05133886402473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.207040309906006, "target_z": 1.9218749963055064, "ce": 6.356205940246582, "genuineness": 4.300237655639648, "identity": 0.4526536166667938, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "e9ff8691ae29cb9298bdb995e9e24b5fa9cf00c229128aba5e08c5c4fbe38ebc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "68062d000301a93b7caf5b63385182291a3c1bfe7a60c82f8c6dc4651c233a1a", "generation_seconds_batch": 21.36141680693254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2380926609039307, "target_z": 0.9609375572646512, "ce": 6.23706579208374, "genuineness": 4.770940780639648, "identity": 0.626107931137085, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/e9ff8691ae29cb9298bdb995e9e24b5fa9cf00c229128aba5e08c5c4fbe38ebc.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "ea2ee885a15af4270e07ebb23940c6494ad938329787c64ce1d34108b1762abe", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c6213bf81c805571a1b195737029574d7847277bc81ed97dba195b012b302a3a", "generation_seconds_batch": 20.832028696895577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.02585506439209, "target_z": 1.7421875942095877, "ce": 6.292664527893066, "genuineness": 4.306513786315918, "identity": 0.5832061767578125, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ea2ee885a15af4270e07ebb23940c6494ad938329787c64ce1d34108b1762abe.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": "ea67f50a88a6b482fbf0dc420184196081929e3c33132e2d50e79e6dd43361a9", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c76114b2c4af9242df78cb0cfb2e4fbe02d68ea6b2e3544f3dc01f67b50ce777", "generation_seconds_batch": 15.341527546057478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.213237762451172, "genuineness": 4.061748504638672, "identity": 0.6562290191650391, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ea67f50a88a6b482fbf0dc420184196081929e3c33132e2d50e79e6dd43361a9.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "ea9a92cb6c23c0b040de22afe311ec4f7197682b80aa3addcd526fab691fbbbe", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d18add07c8c83f9a32b8c8e5ddb13cf398fb432c08a20a94dcca05b9d8e610ac", "generation_seconds_batch": 16.572594539029524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8367919921875, "target_z": 1.554687490763766, "ce": 6.1894097328186035, "genuineness": 4.143336772918701, "identity": 0.6240317225456238, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ea9a92cb6c23c0b040de22afe311ec4f7197682b80aa3addcd526fab691fbbbe.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "eabb72fd2628d4f6844592243227ee0258e43e7073228e6463b5c82cb2376bac", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8fdc3789e72fea5ce66551efb6bdb795e8cb2d2ac7a3307609ce078876a74a32", "generation_seconds_batch": 20.49214577092789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 5.839932441711426, "genuineness": 3.9801597595214844, "identity": 0.7017549872398376, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/eabb72fd2628d4f6844592243227ee0258e43e7073228e6463b5c82cb2376bac.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": "eb33301e470bfaaf0fd219f1bda9befa923a71194b823c4025a402860f877454", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "808bf6633d08768620b3503b820f034a2bf79eb7dfa1fbb87b0cfd9902ecdf98", "generation_seconds_batch": 15.341527546057478, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.268836498260498, "genuineness": 4.105680465698242, "identity": 0.7439457178115845, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/eb33301e470bfaaf0fd219f1bda9befa923a71194b823c4025a402860f877454.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": "eb5cfa35d7c88b94d8c079ebe2483a05d4147c0621d9574b1abfd2d8b27e2157", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f71f02d67d4a60d52923cb927ccc1a48f08002b9cbf6295d5c438c9b1b80706f", "generation_seconds_batch": 21.724717332050204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.046441555023193, "genuineness": 4.444586753845215, "identity": 0.5989280939102173, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/eb5cfa35d7c88b94d8c079ebe2483a05d4147c0621d9574b1abfd2d8b27e2157.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90fda36b37597793a1308f9ace843454a10e90a72b10b55d7afd34c91f69ec1f", "generation_seconds_batch": 16.970804449054413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.568952798843384, "target_z": 1.2890625018472468, "ce": 6.229123115539551, "genuineness": 4.381826400756836, "identity": 0.8572157621383667, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "ebed3b3b2832c03b3ebcb05227a08a74b54059826a82139a8ba4ec00679c4ec1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "198df1365ab366d383e4cb7d119cbb9fa01a2c08b4723305dcab368d8ceb5952", "generation_seconds_batch": 21.39418767497409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.443368911743164, "target_z": 2.156249889165191, "ce": 6.443574905395508, "genuineness": 4.149612903594971, "identity": 0.6151223182678223, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ebed3b3b2832c03b3ebcb05227a08a74b54059826a82139a8ba4ec00679c4ec1.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": "ec3f516c4e22330816827a34c458162a2d9f2f64e000fa005fb2fcd5c386da05", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bf74473628851f094cf36d3ca2932d072413d37212085e410f928747faf2e3d", "generation_seconds_batch": 23.043408458004706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0372133255004883, "target_z": 0.7617188746891601, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.802477240562439, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ec3f516c4e22330816827a34c458162a2d9f2f64e000fa005fb2fcd5c386da05.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "ed3949f010e6f6bb28de483a635b77740970b631f305636e29cd0f785d5a29db", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "63a0b0bbcbc981b8334ce94bb46a65d6be5f9d5a6a3871a3c696d044ac87937a", "generation_seconds_batch": 21.755624048993923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.7208868265151978, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ed3949f010e6f6bb28de483a635b77740970b631f305636e29cd0f785d5a29db.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "ee8e9eccf25f60d3f49e9d7ee25d06fedd112c6120bed31e4de7113ab426f6e1", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 129, "seconds_generated": 10.32, "mp3_sha256": "1dfe94267ff5f428964c43aff2ebf1606ef65ae5a74b141287da1198cabf6521", "generation_seconds_batch": 20.37553729210049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2543060779571533, "target_z": 1.9687500221669618, "ce": 5.990842819213867, "genuineness": 4.143336772918701, "identity": 0.6689160466194153, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ee8e9eccf25f60d3f49e9d7ee25d06fedd112c6120bed31e4de7113ab426f6e1.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ee916edef6a40eee751b30b9bc1750f4b26a70e35edfbce914c17ba31a00a2c0", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "fa7be0d8c72a4c66071c94141dae286f5e6093f9e1f7e71163bbaac58944b1b7", "generation_seconds_batch": 27.232282888959162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5531973838806152, "target_z": 1.2734373355950335, "ce": 6.173524379730225, "genuineness": 4.432034492492676, "identity": 0.7613513469696045, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ee916edef6a40eee751b30b9bc1750f4b26a70e35edfbce914c17ba31a00a2c0.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ee982c077609f57db62d99b6cb2e74684abf246a778021a2319ba7d5e8b486da", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 126, "seconds_generated": 10.08, "mp3_sha256": "1ac3c703cfd4425ed5f09cc5fc91aef1ae1721db64a06de3319ebe61ae0681fc", "generation_seconds_batch": 16.67316015798133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 5.661221981048584, "genuineness": 4.532451152801514, "identity": 0.6754401922225952, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle.", "wer": 0.05263157894736842, "duration_error_s": 0.11999999999999922, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ee982c077609f57db62d99b6cb2e74684abf246a778021a2319ba7d5e8b486da.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "ee9b30546d2ab2df4dc2856734b97eb23bc54ac7a436bbe95465127367a0e419", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2d44050437676b84be5e853bc89c978dd6aab8a85cd8ea9c0fdb020c19b0ef9c", "generation_seconds_batch": 21.605806894949637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.300607204437256, "genuineness": 4.250029563903809, "identity": 0.7173928022384644, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ee9b30546d2ab2df4dc2856734b97eb23bc54ac7a436bbe95465127367a0e419.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "efce3e30dcb103f615b1def752b9196546b63ea54011cfc602676b90f825e04b", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "30f1ab98c052d82be1d36219496dc9c08c4d64a4a9ae0a628a7607e387b45c9c", "generation_seconds_batch": 16.21689886599779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8604249954223633, "target_z": 1.57812512191829, "ce": 6.252951145172119, "genuineness": 4.024092197418213, "identity": 0.244205504655838, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/efce3e30dcb103f615b1def752b9196546b63ea54011cfc602676b90f825e04b.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "effa1441984006a1e49da33960ab08913c7f80f84fb95c2a14f9dd9db4177e73", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "67ba3ff858a1e8813254e58fd7aeccb8fd5daea4a71ac69edc85ac9a28ce1efa", "generation_seconds_batch": 15.889512825058773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.7414167523384094, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/effa1441984006a1e49da33960ab08913c7f80f84fb95c2a14f9dd9db4177e73.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "f0ab2989eef1556ccd8d834f8540d572698a024747e2bf80e1af4d2e5a9a579f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efe826574e2865b371e2ea38d6dc694facb741930d130b332c07d140a1f03cdf", "generation_seconds_batch": 16.79831321898382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.505931854248047, "target_z": 1.2265625461811704, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.879340648651123, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f0ab2989eef1556ccd8d834f8540d572698a024747e2bf80e1af4d2e5a9a579f.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "f0af3fe75ce2b93b9817e087150b65e3443172e39742e66e4d949c579daff46e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c4b8148fb0826e6682c7505575528826d8225abce014a48ee80df3a9ee6d952c", "generation_seconds_batch": 20.725853197975084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.078212261199951, "genuineness": 4.281409740447998, "identity": 0.5778520107269287, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f0af3fe75ce2b93b9817e087150b65e3443172e39742e66e4d949c579daff46e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "f0dc17e9b76c045e6b06a3d867eeb2b490c47d10f668a6832e9a52f5fd5dea39", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5461cc7cacac47d1647abfc627d92771daf7743687aec6bf882c961b13447f0e", "generation_seconds_batch": 22.840981137007475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2696032524108887, "target_z": 0.9921876533214857, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.754241943359375, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f0dc17e9b76c045e6b06a3d867eeb2b490c47d10f668a6832e9a52f5fd5dea39.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "f124c727f1ad729d3120def0e29ef2e5bb2f71ef237400664404a529777c7793", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "87c051ef20205cc44ecd4ac320f2ed65fe004b282181df435f74ac8a341ef7b3", "generation_seconds_batch": 16.764418622944504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0573654174804688, "target_z": 1.7734374538188296, "ce": 6.308549880981445, "genuineness": 4.15588903427124, "identity": 0.21140693128108978, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f124c727f1ad729d3120def0e29ef2e5bb2f71ef237400664404a529777c7793.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "f1de0c4ee7467661057fd379e77ce6887d5ecbc9bf9d3038e306491176d2a32f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "241dec54a6e0aca4182b3d4e40da9660e496e578ce2f1de35504e7a624f57488", "generation_seconds_batch": 23.532899847021326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3803482055664062, "target_z": 2.093750169946707, "ce": 6.419747352600098, "genuineness": 4.457139015197754, "identity": 0.6368470191955566, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f1de0c4ee7467661057fd379e77ce6887d5ecbc9bf9d3038e306491176d2a32f.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcc55965511ea137c26abdf6040939c32540eadd950b865c12070056d4af8004", "generation_seconds_batch": 15.863659417955205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9308652877807617, "target_z": 0.6562500073889873, "ce": 6.165581703186035, "genuineness": 4.557555198669434, "identity": 0.6062902212142944, "hypothesis": "Ton message est arrivé après mon départ, je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.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": "f26ae7f59035333056a8b44cbe1eb8380bd12a1129eb375aa5a20d93ae7bd59e", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7c721ae2ea91c87c50cd9af6a4ddb51ad576b181df8810f377b55b69ea2ecc84", "generation_seconds_batch": 16.65113284403924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 5.879645347595215, "genuineness": 4.432034492492676, "identity": 0.7066044807434082, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f26ae7f59035333056a8b44cbe1eb8380bd12a1129eb375aa5a20d93ae7bd59e.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "f2a3d6aab162d9761d347e4f29640ca80bd0dd66b7205fd27a30a1cd3bf23c48", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2536eb419cae641e2c530aa640ae1a1a6fda3a5e08a8a768305c774cc6a310e4", "generation_seconds_batch": 13.600029004039243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3562569618225098, "target_z": 1.0781250036944936, "ce": 6.054384231567383, "genuineness": 4.432034492492676, "identity": 0.8351937532424927, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f2a3d6aab162d9761d347e4f29640ca80bd0dd66b7205fd27a30a1cd3bf23c48.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "f2b33128e748bd6725e60f29907accdc54ec98995cb476389194f98a51cabfe7", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "cfba4ec203708c13d8cedb67dd191b38345a6aca0d66313a102ee0d9fcdb2c30", "generation_seconds_batch": 20.305099076009355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.742260456085205, "target_z": 1.4609374390408552, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.6749863028526306, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f2b33128e748bd6725e60f29907accdc54ec98995cb476389194f98a51cabfe7.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": "f2e09930e88834a73584f8d283f0c087f7a29e66ec21b390e828dbc4ec6e1226", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "208ebe04b27963a701fe8e955720bbdf346ac76672b5d447813b421317ac12c4", "generation_seconds_batch": 14.491685796994716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.300607204437256, "genuineness": 4.281409740447998, "identity": 0.6321324110031128, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f2e09930e88834a73584f8d283f0c087f7a29e66ec21b390e828dbc4ec6e1226.mp3" }, { "id": "vn_S_FORM_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-0", "key": "f2f4be359172fa1befde92fc0d38ef336e95650aa7fa1cbbe14d3ff8d2ceaf81", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La puerta sigue abierta.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c3282f761004d608684286e231840130ff212c39751bfd9ac0c79144a6637642", "generation_seconds_batch": 21.030747838085517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2464284896850586, "target_z": 1.9609375572646512, "ce": 6.109982967376709, "genuineness": 4.143336772918701, "identity": 0.6511578559875488, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f2f4be359172fa1befde92fc0d38ef336e95650aa7fa1cbbe14d3ff8d2ceaf81.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": "f3cbe034ab520cb20a2974b4cfcab8707f5475c9a0ad6e42abdd3444db6ec771", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1604d851059867d8afc33dc7a97596f9eaff184f0c0e6840d9cdb9ef4f4007c5", "generation_seconds_batch": 16.245029855985194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3173270225524902, "target_z": 2.0312499778330384, "ce": 6.292664527893066, "genuineness": 4.093128204345703, "identity": 0.7470413446426392, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f3cbe034ab520cb20a2974b4cfcab8707f5475c9a0ad6e42abdd3444db6ec771.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "f4596ce87eddb39643d066a88cf2393b56a6262b11f0e1b938bcf5c05baa5818", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": 129, "seconds_generated": 10.32, "mp3_sha256": "25078b7cfb572ec7d11a3e724fa74c5db718fe71f588cbe8c10cccbe081fbe31", "generation_seconds_batch": 16.79831321898382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 6.0305562019348145, "genuineness": 4.15588903427124, "identity": 0.7686644792556763, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f4596ce87eddb39643d066a88cf2393b56a6262b11f0e1b938bcf5c05baa5818.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "f463039b4b5cf42d01ea029f87a2b45b7b2e960d8f2cebf6ddb57ad6142288a3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "25501333717aa6e732949884c9b90d07876bae105a9b1a13a96b273227be8ecb", "generation_seconds_batch": 21.26672684098594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0967535972595215, "target_z": 1.8125000147779746, "ce": 6.149696350097656, "genuineness": 3.8169827461242676, "identity": 0.2988327145576477, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f463039b4b5cf42d01ea029f87a2b45b7b2e960d8f2cebf6ddb57ad6142288a3.mp3" }, { "id": "vn_S_FORM_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-1", "key": "f4735ff2a5de89e6e0d30258568326d831aac03987b679d7cf905dce0ef7993a", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Het stond naast het oude raam.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "75359b82f2f4f5d08aa25a57dc2a17937c83f60eff7ebbd317060b5c20edf634", "generation_seconds_batch": 26.68020153895486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2227954864501953, "target_z": 1.9374999261101273, "ce": 6.229123115539551, "genuineness": 4.293961524963379, "identity": 0.7424034476280212, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f4735ff2a5de89e6e0d30258568326d831aac03987b679d7cf905dce0ef7993a.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "f47860bc18f75e323da75430fcfe6831f96348ae476e3a34306b0e717538bb05", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "56366f660f818233a52c06db48a9dbac1e1055d6c1eda11694e7fc4500c9c8d4", "generation_seconds_batch": 15.391588304075412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.6481510996818542, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f47860bc18f75e323da75430fcfe6831f96348ae476e3a34306b0e717538bb05.mp3" }, { "id": "vn_S_FORM_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-1", "key": "f5002823eee6e3ffb3655883f3f5802f83ff9bf0214483f6aba0f27ec6e6ab51", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "a426e512e1aa8b8584380129f18d8e33111f382d8cb2a7a17524ae0a98958f52", "generation_seconds_batch": 26.06565226300154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.427689552307129, "genuineness": 4.168440818786621, "identity": 0.6209897994995117, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f5002823eee6e3ffb3655883f3f5802f83ff9bf0214483f6aba0f27ec6e6ab51.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "f5c8d91444fd42b6db635b85d8f8473c5f0708530c3b8298f0df428fdf78b94f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "dd418fada24563c972d16962b54df8bfa90faabc4e45675686b37a45beb676ba", "generation_seconds_batch": 22.282496836967766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.923445701599121, "target_z": 1.6406248411367739, "ce": 6.419747352600098, "genuineness": 4.344170093536377, "identity": 0.6913049221038818, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f5c8d91444fd42b6db635b85d8f8473c5f0708530c3b8298f0df428fdf78b94f.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "f65052e86e1889731ba4eb944f4c55df1825001a3c1b0fa382cf937842c23116", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4a02df4aa4bd36e102448996f29a5781a47adb6f230718e92b8decc2261f096c", "generation_seconds_batch": 22.450672763050534, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.252951145172119, "genuineness": 4.218649387359619, "identity": 0.17902550101280212, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f65052e86e1889731ba4eb944f4c55df1825001a3c1b0fa382cf937842c23116.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "f6cbd97e67df13a3d342f06aafabd5597af557f87b228364bb05a9032dd63848", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8ed68a422854ecc66a4b34c83fd6b92ae067c34a3233364d8d18a5101407b201", "generation_seconds_batch": 22.52458139101509, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.7265626644049665, "ce": 6.284721851348877, "genuineness": 4.149612903594971, "identity": 0.28650450706481934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f6cbd97e67df13a3d342f06aafabd5597af557f87b228364bb05a9032dd63848.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "f71f87e8f8f94c47701d095beac034a894bbe2a9321ce4a4a4816bc66d6bf9b4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "864feabf73825d6d61e84acfb4b343ae28b9158f9ca5b8fd26198ceb6e50f995", "generation_seconds_batch": 21.194156382000074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1435611248016357, "target_z": 0.8671875055417404, "ce": 6.403861999511719, "genuineness": 4.20609712600708, "identity": 0.3400697708129883, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f71f87e8f8f94c47701d095beac034a894bbe2a9321ce4a4a4816bc66d6bf9b4.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "f770853243143d3231833e4f34e87ab5a7e8f03df8b3d6a99b9a5fcc2d1d94ee", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ba968fb01c775cc0b3f9d04f83931386f60fc3b00e5450a894b3dc2714bca6d9", "generation_seconds_batch": 21.673605085001327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.078212261199951, "genuineness": 4.287685871124268, "identity": 0.6982970833778381, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f770853243143d3231833e4f34e87ab5a7e8f03df8b3d6a99b9a5fcc2d1d94ee.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "f79930b43e7d91b4f2f77d30e6f0ba3830990ba5ebf8da198148a8bdd9d73090", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "26c306389ef88abfeb90d6ae883e4725f259482975b70befd31e06a1ac824013", "generation_seconds_batch": 22.202818003017455, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8446695804595947, "target_z": 1.5624999556660764, "ce": 6.340320587158203, "genuineness": 4.20609712600708, "identity": 0.40119802951812744, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f79930b43e7d91b4f2f77d30e6f0ba3830990ba5ebf8da198148a8bdd9d73090.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49e3bbf2702b4795bd92c4c0020b52ca14cfc91cb9cf71132b0d8be4280ae664", "generation_seconds_batch": 15.541187877999619, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.994344472885132, "target_z": 1.7109374981527532, "ce": 6.141753673553467, "genuineness": 4.375550270080566, "identity": 0.62845778465271, "hypothesis": "Les clés sont sur la table de la cuisine, j'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.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": "f7c6218e21806a69a620b77aabe1d6fe68d9b27dcebb559fc39373556e079353", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b1e5ca294d5dc2058f3f287b9b41f44138040c0fbbfe803a594f2afd1f542e3", "generation_seconds_batch": 22.597686166060157, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.348837375640869, "target_z": 2.0624998374422803, "ce": 6.308549880981445, "genuineness": 3.9801597595214844, "identity": 0.6894117593765259, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f7c6218e21806a69a620b77aabe1d6fe68d9b27dcebb559fc39373556e079353.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1463b33568c6fef6eebeeeb8e0d305cb20ca1351d870cc7de354241ee5e75e7", "generation_seconds_batch": 14.952156457933597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.316492557525635, "genuineness": 4.231201171875, "identity": 0.6047656536102295, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "f97cbec3acf035a7cb860120fcd56eb1f053aca9569363487f76be251ec509e8", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6731be51cfc169a7d38de37ea7fe26bc76f14ee646c6f08a96e153e952548be9", "generation_seconds_batch": 21.063573995023035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.332377910614014, "genuineness": 4.068024158477783, "identity": 0.348244309425354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/f97cbec3acf035a7cb860120fcd56eb1f053aca9569363487f76be251ec509e8.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "fa0ebd53e6612f6965166e932a05f8091d50611074d748735befa1dc15fd9e24", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "746f51419eb655fdb3fcfed0c4bf53b35c217c00a7938608a62c4c9cfd5fc413", "generation_seconds_batch": 22.569629327044822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7343828678131104, "target_z": 1.4531249741385446, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.24539656937122345, "hypothesis": "I found the little box. It was beside the old window. We can look inside", "wer": 0.0625, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fa0ebd53e6612f6965166e932a05f8091d50611074d748735befa1dc15fd9e24.mp3" }, { "id": "vn_S_FORM_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-0", "key": "fa299546ae5fd95a52a4a562850f014108b570df58aed48ac0ccef54f5001edb", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The door is still open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I put your coat on the chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "ba6144d799709d480e856af7ba65d7e7bab0ef385952795748cfa7a32b74b490", "generation_seconds_batch": 21.996487003052607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.133810997009277, "genuineness": 4.017816066741943, "identity": 0.31787827610969543, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fa299546ae5fd95a52a4a562850f014108b570df58aed48ac0ccef54f5001edb.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "fa46d27649cf7efd8117105fb6056ed0e8f8f02d25e3825e791e14880f95bf5c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9bfde6dc14fc5515d38c0f1d165d4f93da6c506d52302dbb2074e48fc30ca7ec", "generation_seconds_batch": 21.422150991042145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.181467056274414, "genuineness": 4.143336772918701, "identity": 0.6512894630432129, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fa46d27649cf7efd8117105fb6056ed0e8f8f02d25e3825e791e14880f95bf5c.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "fa6950c02670c5ea278510e17fe62ec786060ec6d66b6a3dfbe4b2e917d645de", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f9452891d7a13bacd9babeb8dc5855eff88b83c526ba4a36607771f62c9e42ab", "generation_seconds_batch": 21.159915798925795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.64772891998291, "target_z": 1.3671873873179443, "ce": 6.292664527893066, "genuineness": 4.331617832183838, "identity": 0.369892418384552, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fa6950c02670c5ea278510e17fe62ec786060ec6d66b6a3dfbe4b2e917d645de.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "fa8837ea72c2492336576dbf68cc0047b0bd33d12fbfef4212ebbf8d76f8009c", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bf81f9aa4d5560e4f574b63705ea08a4c19e3326db09ef9c930ae949f057d565", "generation_seconds_batch": 23.337437011068687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.136141777038574, "target_z": 1.8515625757371195, "ce": 6.284721851348877, "genuineness": 4.18099308013916, "identity": 0.2932516932487488, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fa8837ea72c2492336576dbf68cc0047b0bd33d12fbfef4212ebbf8d76f8009c.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "fb69f7ecf511cd3f7999a73b1964e76fdf1962945df65f264c9303121b486810", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e390d7f04bbb04af4dca1f32d3a9f632f65a1413818b3cd7304f136e11f6dc1c", "generation_seconds_batch": 21.916815120959654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.109982967376709, "genuineness": 4.13078498840332, "identity": 0.7832467555999756, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fb69f7ecf511cd3f7999a73b1964e76fdf1962945df65f264c9303121b486810.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "fb6cf8ad96a150a79820cf8e0183a5c1c87e30363bed54f41edc1502d827dcbc", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "059c9e2c917c8786ff174de6a443e47e03e67798023b6d5ceeaab19ec6631184", "generation_seconds_batch": 21.78842439490836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4586660861968994, "target_z": 1.179687520319715, "ce": 6.300607204437256, "genuineness": 4.306513786315918, "identity": 0.41181591153144836, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fb6cf8ad96a150a79820cf8e0183a5c1c87e30363bed54f41edc1502d827dcbc.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "fc1ff961ec9f49314fb2f59f4aec903b0190ea09218d4b3b852e3279c2a9fe76", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ae7d45b974375610303e7498f5f7476e7d1ff780a173c847798129bee43cb3e1", "generation_seconds_batch": 21.05324939393904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0923566818237305, "target_z": 0.816406365452926, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.7633221745491028, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fc1ff961ec9f49314fb2f59f4aec903b0190ea09218d4b3b852e3279c2a9fe76.mp3" }, { "id": "vn_S_FORM_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-0", "key": "fca5bc4fb55d8dbeef2fb5ae04b20029b84d6f61d61c118bec90b7d192cc19e5", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] La porte est encore ouverte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4bbf95dae97b1bf1693a8e36f01f9ec04ab3e08f96866bf52c2a1cc45b3d2f2a", "generation_seconds_batch": 20.813051688019186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.205295085906982, "genuineness": 4.432034492492676, "identity": 0.6320459842681885, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fca5bc4fb55d8dbeef2fb5ae04b20029b84d6f61d61c118bec90b7d192cc19e5.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": "fcb39118b44feddafa07c042901c6ab78e1aa819fa15db579a2ff7f6b43f8ac5", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] The keys are on the kitchen table.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "49bca2e1a5c4f445cb107fa69bea96f9d5d710e1c5145749252147c9636825a4", "generation_seconds_batch": 16.10862778907176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0100998878479004, "target_z": 1.7265626644049665, "ce": 6.3641486167907715, "genuineness": 4.055472373962402, "identity": 0.31552767753601074, "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": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fcb39118b44feddafa07c042901c6ab78e1aa819fa15db579a2ff7f6b43f8ac5.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "fcde562217d25ad257b48eee14dc1bc284db36a79203e5f0d81b72428beb39ab", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f65a40c9b7be5fcf70dbd9f377f91cb2c01ab3890ee9ade193c935f7a96aa209", "generation_seconds_batch": 21.868054853053764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.781648635864258, "target_z": 1.5, "ce": 6.260893821716309, "genuineness": 4.406930446624756, "identity": 0.6550891399383545, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fcde562217d25ad257b48eee14dc1bc284db36a79203e5f0d81b72428beb39ab.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "fd25a60ae21d46631fe59d2a7f7a2d12c44ca73bbd7a8539b3dfb4c6665fd67f", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a96d98f5d21ddfc272d25728d71352becf4716dd254beee02be2a98125e62d93", "generation_seconds_batch": 24.388659139047377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.660513162612915, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fd25a60ae21d46631fe59d2a7f7a2d12c44ca73bbd7a8539b3dfb4c6665fd67f.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "fd69b26ff72067f282b42d2f3d4ee32f8c3b3edbca116610b484c121487e991d", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "eaff5473c5e80561e0e57812235764ea4e51dd30ea4a69752eca35f1d1218d79", "generation_seconds_batch": 23.593673208961263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.117925643920898, "genuineness": 4.306513786315918, "identity": 0.6530410647392273, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fd69b26ff72067f282b42d2f3d4ee32f8c3b3edbca116610b484c121487e991d.mp3" }, { "id": "vn_S_FORM_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-de-0", "key": "fda3da51cb7203e0a67e56f117c459a1fc0c994b0bcb77624431466f49b426e3", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Die Tür ist noch offen.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "8a6d6add3e79740fa52b78828e2412492f64b5cd73f9687a0f2df20d703ffee9", "generation_seconds_batch": 23.337437011068687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9707117080688477, "target_z": 1.6875001034458217, "ce": 6.229123115539551, "genuineness": 4.369274139404297, "identity": 0.6932363510131836, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fda3da51cb7203e0a67e56f117c459a1fc0c994b0bcb77624431466f49b426e3.mp3" }, { "id": "vn_S_FORM_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-en-1", "key": "fe023a4817179c4e758a16a3b053644a0e34c1830a0645e79240acee8a936325", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] I found the little box.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It was beside the old window.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "972e150e4bcc6cc2cd7a53b049d0c8d9d6912168eaac3755248d599d03be1756", "generation_seconds_batch": 20.162616575951688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9076905250549316, "target_z": 1.624999911332153, "ce": 6.38003396987915, "genuineness": 4.369274139404297, "identity": 0.2836116552352905, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fe023a4817179c4e758a16a3b053644a0e34c1830a0645e79240acee8a936325.mp3" }, { "id": "vn_S_FORM_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-nl-0", "key": "fe524df7a1b5e39f3d45f90d718e40fd5c1a0c8fafa67bfae763b15719aae00e", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] De deur staat nog open.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1fc274c20eda3a22095d339e5b16d5530820aac30d11856f600e48cac4dc1fd", "generation_seconds_batch": 24.828868657001294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4192779064178467, "target_z": 1.14062495936057, "ce": 6.09409761428833, "genuineness": 4.262581825256348, "identity": 0.7405174374580383, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/fe524df7a1b5e39f3d45f90d718e40fd5c1a0c8fafa67bfae763b15719aae00e.mp3" }, { "id": "vn_S_FORM_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-fr-1", "key": "ff59f25abffa5a374eedd88447146224bd9d1dbe802cde4ba720442582e070d4", "bucket": "vn_S_FORM_reg", "phase": "grid", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3783b31bc8db214fc97780e8e6b0cf77291b7a38633f40a2e0b9c361a3214370", "generation_seconds_batch": 21.543968473910354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8840577602386475, "target_z": 1.6015625166252214, "ce": 6.1020402908325195, "genuineness": 4.331617832183838, "identity": 0.7731364369392395, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ff59f25abffa5a374eedd88447146224bd9d1dbe802cde4ba720442582e070d4.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": "ffacf7be2803b435f2f6acc36d3217da4760c77b32dcb3f6a9e6083eb7c9d381", "bucket": "vn_S_FORM_reg", "phase": "heldout", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Someone moved the little blue chair.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "801ce9611ed89451b214e653aa99652cb462eac32a3a20aaf3aa42deca013320", "generation_seconds_batch": 22.32076741498895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.080998420715332, "target_z": 1.7968750849733535, "ce": 6.507116317749023, "genuineness": 4.218649387359619, "identity": 0.39018186926841736, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ffacf7be2803b435f2f6acc36d3217da4760c77b32dcb3f6a9e6083eb7c9d381.mp3" }, { "id": "vn_S_FORM_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_FORM_reg-es-1", "key": "ffcc3a1539b93d6a39821a0fc0817d2ffe9a35512bf9abd36b851c04017e866c", "bucket": "vn_S_FORM_reg", "phase": "wording", "candidate": "manual-b9f763b6bfff", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.\nSCRIPT:\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Establish it immediately. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] He encontrado la cajita.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Sustain the delivery. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Formal, composed public speech; precise articulation, measured rhythm and professional restraint. Preserve the intended style through the end. Begin with the feeling held back, then let it leak through and become unmistakable.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "244d33a2cd9af79fe9eceff33138225e577d8821370c86210b23e6baf54e26f2", "generation_seconds_batch": 16.605416303034872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.173524379730225, "genuineness": 4.557555198669434, "identity": 0.6921594738960266, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_FORM_reg/ffcc3a1539b93d6a39821a0fc0817d2ffe9a35512bf9abd36b851c04017e866c.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "48d3aa5197cfc7931d5df384b48c81de259c6c95f7e83ff46e72d7add854314f", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4901766777038574, "target_z": 1.2109376163765495, "ce": 6.054384231567383, "genuineness": 4.482243061065674, "identity": 0.5509204268455505, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/62458816db120582dba73400.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "4875f47c208b946ee7f42f89493130fff4b648ac67f2c2a5300cfa297fc7f078", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.986466884613037, "target_z": 1.7031250332504426, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.4252471327781677, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/483cfb7223a2d1e724348824.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "ad04b0c55cfd927acb2a421589f7b56a82ca9c4296e144b7822dfb098fa4c1b6", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0337324142456055, "target_z": 1.7499998226643056, "ce": 6.157639026641846, "genuineness": 4.2249250411987305, "identity": 0.5273939371109009, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/fbc186f4dd13696faad09a6c.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "37403df0605cbc5c7111b6685168bb6cfc1ac975a24d67e3a16e87a90bb4494b", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.144019365310669, "target_z": 1.85937504063943, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.3649364709854126, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/1da44adfcb84fd2351047c17.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "723d6681770fa9e4015e0877d8f7cb8b8afe266436e83ee6d9ece4e437134011", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.737863540649414, "target_z": 0.46484378971580653, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.40946245193481445, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/f0375c2417a1cca148bfa5b1.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f0632fd9c11a996971b4a138b41741f739638c837f278dd05973c6c9635a81ea", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0652432441711426, "target_z": 1.7812501551687325, "ce": 6.23706579208374, "genuineness": 4.218649387359619, "identity": 0.4013039469718933, "hypothesis": "Die Tür ist doch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/d51cd37dd659af95ff3990d8.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "af7431c0e43b4cd92ce787743c07a82b1fa085d7c0b728d2f39948b1310bca42", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.411400318145752, "target_z": 1.1328124944582596, "ce": 6.1020402908325195, "genuineness": 4.494794845581055, "identity": 0.5593881011009216, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/0a0c8473fa2d64593237b2ef.mp3" }, { "key": "original-vn_S_FORM_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "49a825c3d4cb9cace61c6aabc48e7ba0901e6661f926787ac136ba7e5ab335cb", "task_id": "vn_S_FORM_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710750102996826, "target_z": 1.429687579431613, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.6422777771949768, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/496e6c01441323adb29fd780.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "e696efa823375fdcef271de0f96b620532f68ac851bfdfa62b980450769f1375", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1518969535827637, "target_z": 1.8671875055417404, "ce": 6.340320587158203, "genuineness": 4.193545341491699, "identity": 0.3312576711177826, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/cfa42ad33ecc75c2a1e49bad.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "023295aa8eb11f8cf636319e3e767376842814c66ca2732e9b4c5af8077a0cc5", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 6.38797664642334, "genuineness": 4.287685871124268, "identity": 0.662835955619812, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/ccd6fb460dbd3bb7458706f7.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "67f48dad21be4b145beac090bb6b3bcde9cfb9bca6fcedbd3fac61b2f9d4444d", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3645925521850586, "target_z": 2.078124767246901, "ce": 6.395919322967529, "genuineness": 4.231201171875, "identity": 0.6959473490715027, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/dc06e1b59072644b38765ed6.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "020744c8eb46d714db2182dc17368a9ac53a5cad70d717d46e233b9f844b3c6b", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1282639503479004, "target_z": 1.8437498743872165, "ce": 6.356205940246582, "genuineness": 4.337893962860107, "identity": 0.5567379593849182, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/ad2ab513b43a35109753d349.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "1a07a83c5074bc8e5df66704e95f7fbfe3e87171d48f3852ee4018b53a6e42e5", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.356205940246582, "genuineness": 4.218649387359619, "identity": 0.4810953736305237, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/dadb9dd80b9bb420cdab98f0.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "d27007907f0d62354f4484ac4df1e5600fc67e954798ffa0f75fd73c5899d454", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2385506629943848, "target_z": 1.9531248559147483, "ce": 6.538887023925781, "genuineness": 4.457139015197754, "identity": 0.5326786637306213, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/ca45b7251c6e2111297fed6a.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "0b199b7c9084e0d7500bf3825ca3487b68653ad813d8b9ed15e682117d742080", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301571846008301, "target_z": 2.0156250480284172, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.586025059223175, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/5c65a6f84308d558bfb03bae.mp3" }, { "key": "original-vn_S_FORM_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "326364205d6a22e195a4075cb62ea349155e4ab0c13ab19b9ba7bcb95ec31f56", "task_id": "vn_S_FORM_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.50639009475708, "target_z": 2.21875008127886, "ce": 6.2767791748046875, "genuineness": 4.049196243286133, "identity": 0.36247068643569946, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/de/4f12df2d5542faecf9c6fe23.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f71c34e4a211f1b4a183a76bc0d76ee23cc7643118944cedd2878325cefdd7e2", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6713619232177734, "target_z": 1.3906250184724682, "ce": 6.292664527893066, "genuineness": 4.193545341491699, "identity": 0.2373623102903366, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/21f66d264eaf578be8c84dba.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7500feb7fd5fd7fdd775e66c4c38253fc1e29de8fca3bcbf8fdae7f7c9a8bb16", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5768303871154785, "target_z": 1.2968749667495574, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.2948727309703827, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/62901ede2d07e00b9f002132.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "42a983721f5999c543eab9bbc63d070e2a0b5f37468c8e4b950569c63ea7066c", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.348379373550415, "target_z": 1.0703125387921832, "ce": 6.268836498260498, "genuineness": 4.2751336097717285, "identity": 0.3679337799549103, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/bc41e0c52a2e2d84968ad02e.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "93f41f83ae1fe2090c2763a79da8c66ad4a2534e85439f3a24a7c87dd3511b2c", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6949949264526367, "target_z": 1.4140626496269921, "ce": 6.268836498260498, "genuineness": 3.8295347690582275, "identity": 0.15448340773582458, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/fa93fc03d6b52b1b6a30a1d0.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "fa7e0c6c53c6ebb8ef44aeb7c840e456b41defb5bcd4e78bcc97a22e5ab2d49e", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.954956531524658, "target_z": 1.6718751736412008, "ce": 6.054384231567383, "genuineness": 3.8295347690582275, "identity": 0.29553544521331787, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/a139164e5f7e1b982ce3ef29.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "df8b8a19f2cf2a6ec3e7328882176f9c35c662ced14019bbfc140f6d757e5873", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.262183666229248, "target_z": 1.9765624870692722, "ce": 6.221180438995361, "genuineness": 3.722842216491699, "identity": 0.08965595811605453, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/9c163381c445170a6eaddf5b.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b5b71d8d42eabe8b01ca0bba3877f77135472a049927c26be22ed86bc7c8ac30", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 1.1484374242628805, "ce": 6.078212261199951, "genuineness": 3.9801597595214844, "identity": 0.1403358280658722, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/dc6b90ee429f92b7c6017497.mp3" }, { "key": "original-vn_S_FORM_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b1b20af81f2caaa0a88e5231a073a95c0b3157c0ce2b08ba87ae64739b36b383", "task_id": "vn_S_FORM_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.655606746673584, "target_z": 1.375000088667847, "ce": 6.292664527893066, "genuineness": 4.18726921081543, "identity": 0.301613450050354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/ffb39afc4d13835a288d54cb.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "76184e41ea485e8aeaa3e547bfca657140a55e3b37ad9f81536a753751de29bf", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7043836116790771, "target_z": 0.4316406365452926, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.3347567319869995, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/9f13f311dbf0c6bd888b69c9.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "c813b4328c3c594ceb5cc3dc5bacd3136d7bc3cc2c7fa9e31e3f5b79bc5ef630", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9702534675598145, "target_z": 0.6953125683481322, "ce": 6.324435234069824, "genuineness": 4.369274139404297, "identity": 0.5021659135818481, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/5cd01ce207575bd5857e71dd.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7d8b659fc338d94d9a5b8fe340ccb3957eff8e5f8a05a323e1055410f21b20e9", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9151101112365723, "target_z": 0.6406250775843663, "ce": 6.324435234069824, "genuineness": 4.381826400756836, "identity": 0.303474098443985, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/a7d048d2127d66ed56f17fcd.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7ff023af0de9ac4b76d9d607a2adc355b31734d6db9d84ec29dd246eaa807f15", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8717830181121826, "target_z": 0.5976561659502698, "ce": 6.332377910614014, "genuineness": 4.469690799713135, "identity": 0.4096085727214813, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/cc71dc7025162001a296ba8a.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "2a487f9e019cd4e7287cd619dd9be0f66b3cf29d31a1fa1c6e3350144308679f", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8835995197296143, "target_z": 0.6093749815275319, "ce": 6.3641486167907715, "genuineness": 4.595211505889893, "identity": 0.5056486129760742, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/f935e4bc1c2656e968513aaa.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "e037179f34a52f4780701a6a5c47e0b61243765fb070f102b55d6b2a4643c903", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1475000381469727, "target_z": 0.8710938562166919, "ce": 6.38003396987915, "genuineness": 4.319066047668457, "identity": 0.27982795238494873, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/0370a5f796898dc402d37692.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "561ba92a5f34ca343fae628e1c26ff178146369b0c0a3f6b19de8785e7c61339", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9466204643249512, "target_z": 0.6718749371936082, "ce": 6.419747352600098, "genuineness": 4.507347106933594, "identity": 0.2634875178337097, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/f1cf3a5cc3052463cacc1ea0.mp3" }, { "key": "original-vn_S_FORM_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "103b78c7a2413e879e5eeb917cfa32ad6bece4eaff2f15562fcd17e6169d0f07", "task_id": "vn_S_FORM_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8678443431854248, "target_z": 0.5937500517229108, "ce": 6.3641486167907715, "genuineness": 4.457139015197754, "identity": 0.26754164695739746, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/en/15af9965d5b74b5aec42a6a1.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "aa53dfeb9e1c2e90c2f1bdd8e45e1da41e48855e952403408e5a483356a277df", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5221452713012695, "target_z": 2.234375011083481, "ce": 5.998785495758057, "genuineness": 3.8295347690582275, "identity": 0.5932515859603882, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/d538089e7c58713f6e9f4362.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "03507792439067ff20aa59b2f4b15a44f6843a151109d9cb1ce4e454bddfed55", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.078212261199951, "genuineness": 4.13078498840332, "identity": 0.6243016719818115, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/134de28be9ab89ef729cc3d4.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "50f8f4421307fc0c9eeb27a728343ffe94da460f0f29ff253e4ff42280797524", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1518969535827637, "target_z": 1.8671875055417404, "ce": 5.998785495758057, "genuineness": 4.143336772918701, "identity": 0.7141748666763306, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/27b2cd2ba0892458650d91f1.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "68e3d323780a4f83e66dd2eed13da093b0330ea83591f250d91fc38e6fa88d89", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112508773803711, "target_z": 1.8281249445825956, "ce": 6.006728172302246, "genuineness": 4.042920112609863, "identity": 0.6302741765975952, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/08bea89f26914f063776533c.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "68a962df562734478f6af0e4d1f2c63849fa15c44946269cbc48c9526dcffae8", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.805281639099121, "target_z": 1.523437631154524, "ce": 6.2767791748046875, "genuineness": 4.519899368286133, "identity": 0.4652097225189209, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/d6b66617a9007f972a588333.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f1ef968f8680a46ca9e75b5dc6dfc79c63009023b9b045832008aa1ca9cc81a5", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3961033821105957, "target_z": 2.109375099751328, "ce": 5.982900142669678, "genuineness": 3.9927117824554443, "identity": 0.5384992957115173, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/aa0c15f6f901349a6a427a93.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "08b4fd1a61691e82b354947ad1892405b8d5a0e3cfe1600ed66127e92ebf8782", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2696032524108887, "target_z": 0.9921876533214857, "ce": 5.839932441711426, "genuineness": 4.319066047668457, "identity": 0.6104505062103271, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/db3b87bd07102a82bfff8449.mp3" }, { "key": "original-vn_S_FORM_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b2088189b16e55073a54b13a7dd6148ab43b17ad74d38c4f2073b7ecdf018b2d", "task_id": "vn_S_FORM_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8131589889526367, "target_z": 1.5312498596092419, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.5498671531677246, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/30ac13ac95f10524a51617fe.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "8304759105b0ab48e4378bb0851d870cb59fda77fd7fccdd10707cf33a9cc27b", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1597745418548584, "target_z": 1.874999970444051, "ce": 5.935244083404541, "genuineness": 4.086852550506592, "identity": 0.7396137714385986, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/808be94d776e02f48ef2df44.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "94086b1c36d4bac0ba30a34370bcf5988ee12274a28cce56f1523fde0330dc42", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.16719388961792, "target_z": 0.8906249002486719, "ce": 6.1020402908325195, "genuineness": 4.6956281661987305, "identity": 0.7348507046699524, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/d28aca4686d1692453141b8c.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "462d7b6c86167f97e4f7deacd82d25cebd4add66bc8755f2cf812480be80ff47", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.8052419424057007, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/3f9c52a0b1572f29068699f5.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "71a4471ea842f69eeeacb87fe0b497a686772c106165d02811f4fd3279175e78", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 6.252951145172119, "genuineness": 4.645420074462891, "identity": 0.6444656848907471, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/0414bb92531d980ccc0b690c.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "e6edc1cf7aa79d49545979d1e352280e4a84ee24df4835e690eadbd84fee1473", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.474421262741089, "target_z": 1.195312450124336, "ce": 5.982900142669678, "genuineness": 4.570107460021973, "identity": 0.7562985420227051, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/a0d4c54fefce9cdbefe42500.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "764e80f80be0f647df03c0949e23ebf72feefe8123ecc039a33bdd87f1d3be95", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4271554946899414, "target_z": 1.1484374242628805, "ce": 5.95112943649292, "genuineness": 4.532451152801514, "identity": 0.6528224945068359, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/08b38eb33ccf837baac2fe1c.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b6d04a3a8480c3edb4db8728ecfce62b9834eb71e5a05ee6e4d87908b0508fdb", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2459702491760254, "target_z": 0.9687500221669618, "ce": 6.133810997009277, "genuineness": 4.7332844734191895, "identity": 0.6365936398506165, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/5f0da7e62f81257923e1066d.mp3" }, { "key": "original-vn_S_FORM_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "cae5eda1caac6e55224041b6a6ac4414684378c1f74fc20dad09f411d0c02019", "task_id": "vn_S_FORM_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 5.998785495758057, "genuineness": 4.557555198669434, "identity": 0.7740354537963867, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/es/2b8adbae78789da3c179bb2d.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "fc197b20b545774860b3f4411d5a92ab24356f334c14e2e8390760b6341267a3", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.163255214691162, "target_z": 0.8867187860213129, "ce": 6.165581703186035, "genuineness": 4.457139015197754, "identity": 0.8211727142333984, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/76a416a8505a24248aa58432.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f4707d8f02aba966febc80729ea4989b7a9ffeaf132b4f313e623b521cd3a1b9", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1199283599853516, "target_z": 0.843750110834809, "ce": 6.09409761428833, "genuineness": 4.582659721374512, "identity": 0.7288832664489746, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/c3e3b9ca2c1d9c5c95cbe58d.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f490dda3277732126f5d85d280886934fc4d3ac05ab2e982fd6fac56e899a45a", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.372012138366699, "target_z": 1.0937499334991145, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.5872690677642822, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/255a087b6f13f04297c951ac.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "9d9f045ed63aece9e0168ca026d8b518d1f8ae6a844d5f26c79a1b4630e3f368", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8761801719665527, "target_z": 1.5937500517229108, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.7479544878005981, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/794d42bcd6316d0de29168f0.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7877740cdb5b2bc862bf7a1de626a03b34aee993bef69191260e2b66bdcc8768", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5216870307922363, "target_z": 1.2421874759857914, "ce": 6.125868320465088, "genuineness": 4.507347106933594, "identity": 0.7203531265258789, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/50f1d914859a760b060f604e.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "6d1d29b6ba22804f3db44bfbc732205c101165c5976166756d5c92f6fb93f1d4", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.805281639099121, "target_z": 1.523437631154524, "ce": 6.133810997009277, "genuineness": 4.400654315948486, "identity": 0.7769373059272766, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/3bd7ca447f8ddf25bcd2288c.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7813151bea42a0eb82c2e1f41ebbba23d862d3775f12e058a8dec050517bebb7", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4035227298736572, "target_z": 1.125000029555949, "ce": 5.911416053771973, "genuineness": 4.18099308013916, "identity": 0.46000444889068604, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/3c373387b60834b2d01bb1cf.mp3" }, { "key": "original-vn_S_FORM_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "9d2d785f1efaec61005520e442ba5591b57264883b51d899fcb09d1a7c1aecc8", "task_id": "vn_S_FORM_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5138094425201416, "target_z": 1.2343750110834808, "ce": 6.022613525390625, "genuineness": 4.325342178344727, "identity": 0.5509998202323914, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/cc8ad3455b31067d817aff22.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f12a6faed5efba52fc2db07d7b8f6c39165b9ffdcd492daf69ff6e103c3a73e8", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6083407402038574, "target_z": 1.3281248263587992, "ce": 6.046441555023193, "genuineness": 4.381826400756836, "identity": 0.7876325249671936, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/175136595da8b168bab620db.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b01554d6fd6dbe43f8c0908b8944d185a24472ed6eb92305e68905393679c39e", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3405017852783203, "target_z": 1.0625000738898727, "ce": 6.197352409362793, "genuineness": 4.469690799713135, "identity": 0.6616024971008301, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/48fed14b9aedffe7865d7fe5.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f6f12351aec8a855116a83ff7d61f42332fdaeb413ba76853e605c86dacefa74", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4822988510131836, "target_z": 1.2031249150266465, "ce": 6.181467056274414, "genuineness": 4.519899368286133, "identity": 0.7365080118179321, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/c84f0f68435c544dbe4edac9.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "df429583f3984e3ad746fd03cc357541ee14290345f80f20018cfdf3cc0508e0", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3877673149108887, "target_z": 1.1093748633037357, "ce": 6.181467056274414, "genuineness": 4.582659721374512, "identity": 0.6811895966529846, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/7d5c0e6d756a8616b6b17283.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "e5aef5943494eb55a05e6cf9a5659792a55f8a6e01976cc884c6067bbed3ef04", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3330821990966797, "target_z": 2.046874907637659, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.7546690702438354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/3bab41720e83abab19517d81.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "9297612911ed6b2e7b57c8a963c76a4d177717a1b4c7eeb7b5c424c0ade6d194", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1868879795074463, "target_z": 0.9101561807282444, "ce": 6.173524379730225, "genuineness": 4.7081804275512695, "identity": 0.7913029193878174, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/bbe8b76b5c98a5f85a64b9e8.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "b37063c4506f59b7bfcbc6cac940f2bbc61ee834b439acb6b8e4da915757005d", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4350333213806152, "target_z": 1.1562501256127835, "ce": 6.133810997009277, "genuineness": 4.293961524963379, "identity": 0.8578920364379883, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/41c5250189a34495aaa6fc2c.mp3" }, { "key": "original-vn_S_FORM_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "15528bd981b62501a5c801a3e682c924cd263acca960fc771f8d5f54afc1fad2", "task_id": "vn_S_FORM_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.687117099761963, "target_z": 1.4062499482770892, "ce": 6.221180438995361, "genuineness": 4.582659721374512, "identity": 0.6501762866973877, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/fr/606aa30aa623b67d1e612ea6.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "dfe71698d2a5c0832f01ca24f5f821bcc0821cbb2b4d93e9a9d7df36c78c56a1", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0253968238830566, "target_z": 0.7500000591118982, "ce": 6.1894097328186035, "genuineness": 4.362998008728027, "identity": 0.48835885524749756, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/cc7e4c6e87f925d22e16e3e3.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "eb595a1d75c287be3733a1d930e1731ad785f9d58b8c8e943d26845224d62082", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5980356931686401, "target_z": 0.326171887468916, "ce": 6.0305562019348145, "genuineness": 4.20609712600708, "identity": 0.7577723264694214, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/39f574d37174f88d98c18d96.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "afa8acc839021853c6405de30312ac434191724d37a184e79aac56ad33409c96", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1435611248016357, "target_z": 0.8671875055417404, "ce": 6.229123115539551, "genuineness": 4.15588903427124, "identity": 0.5294468402862549, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/de92027c24819c071a07ef61.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "5bd4fd281dc671e44c068f882b5f10885c57ffaa91eac0c8e78f2683929acc8f", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1829493045806885, "target_z": 0.9062500665008854, "ce": 6.197352409362793, "genuineness": 4.243753433227539, "identity": 0.7245818376541138, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/2b8835ba2c53af94605ef0b9.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "85ae7a4913a4d423edf29681852433ec4b5546dbf0eb3a463d9eba1813146127", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7028722763061523, "target_z": 1.42187487808171, "ce": 6.173524379730225, "genuineness": 4.325342178344727, "identity": 0.7186875343322754, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/6dcec9902f91c6a91f31aba4.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "22d99237241141cc74ab87c02fa5e7ac4c5ba397be472b8d428b24f882a2e02c", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2183985710144043, "target_z": 0.9414062767850788, "ce": 6.046441555023193, "genuineness": 4.13078498840332, "identity": 0.7133384943008423, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/72709f547a7b4cc02267391b.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "38fc865e84e65fd622cb1e29b49a754a4e9ce1a8a374339836cd1bea5b12b077", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.25384783744812, "target_z": 0.9765624870692723, "ce": 5.998785495758057, "genuineness": 4.20609712600708, "identity": 0.7314808368682861, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/df760c061a7c3d5337ce352b.mp3" }, { "key": "original-vn_S_FORM_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "92240a58c1c4c62086a8c1733474990ba9649dadac6e5ca2acad5e00af0772d2", "task_id": "vn_S_FORM_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891935348510742, "target_z": 1.6093749815275318, "ce": 6.0146708488464355, "genuineness": 4.243753433227539, "identity": 0.6148257255554199, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/40f4d191d5a2ae2d6ed1cc13.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "56d71c0dae324fcc2245886033087f5338653e8424c2b91c4ac471beb3d1aeb2", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624096155166626, "target_z": 1.3437499926110128, "ce": 6.165581703186035, "genuineness": 4.281409740447998, "identity": 0.6345929503440857, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/d5fe7e32fe1c06dd49643bf5.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "5915e92a7ca80745d272ba730f3340c6a1f4396998cd6e82265b4d73c4c9c298", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135683536529541, "target_z": 0.8593750406394299, "ce": 6.125868320465088, "genuineness": 4.262581825256348, "identity": 0.6569908857345581, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/01a0b8451cb3aa69f9a7a4dc.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "43894c6cb396be5a75e5b24086e56398783cd001b271bf602c238ada53232c25", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718627691268921, "target_z": 1.4375000443339236, "ce": 6.09409761428833, "genuineness": 4.174716949462891, "identity": 0.6826872825622559, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/7bf63fc5593cd4c35a010741.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "7eecd3246876fcd4c5558c87f54497fd5c327c0bb420c5fcf74d2110187be593", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9392011165618896, "target_z": 1.6562500073889872, "ce": 6.229123115539551, "genuineness": 4.306513786315918, "identity": 0.58757483959198, "hypothesis": "ik heb het kleine doosje gevonden het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/a94236d9b451f0bd1e441653.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "66aa4fea991c83b16f7c12acf77a9858617fa9dccfcfd3847ea90fadc205e4e9", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.633484959602356, "target_z": 0.3613280977531095, "ce": 6.1894097328186035, "genuineness": 4.632867813110352, "identity": 0.6643699407577515, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/3e82c95a56e086105ebb5dca.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "f952069b55debc55c6d4728447a8685c03a1eb6d4e0b8ab401d33f023e4cd82f", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.710750102996826, "target_z": 1.429687579431613, "ce": 6.0305562019348145, "genuineness": 4.231201171875, "identity": 0.7121509313583374, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/e4e5befb0f3d84c5dcabbfa3.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "27624f15a2dbcfe4bec3b2c6492e2841fe45acc9ac20f042869576979b5dd876", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.285358428955078, "target_z": 1.0078125831261067, "ce": 6.125868320465088, "genuineness": 4.319066047668457, "identity": 0.599662184715271, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/8775789477a7a83f3856ae5e.mp3" }, { "key": "original-vn_S_FORM_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_FORM_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "dbcc1fe41d9885ccb981663c580becaa6147f7bf9d5558a0adc4dbd53570b287", "mp3_sha256": "8968549bffa45fee1eb7b5a1bf3b78309762ba9a9ba4530b6ecb47b06157c5c4", "task_id": "vn_S_FORM_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong Formal Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7265052795410156, "target_z": 1.445312509236234, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.7144010663032532, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_FORM_reg/nl/7cc35d1c35a63c65d1bfaa4d.mp3" } ], "source_standard_audio_unchanged": true }