{"id":"W4417001668","doi":"10.48550/arxiv.2512.02201","title":"Swivuriso: The South African Next Voices Multilingual Speech Dataset","year":2025,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Pretoria; International Development Research Centre; Nvidia; Bill and Melinda Gates Foundation","keywords":"Benchmarking; Domain (mathematical analysis); Baseline (sea); Data collection; Speech technology; Computational linguistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001063516,0.001007281,0.0006171477,0.001923949,0.001036412,0.0009084602,0.00100108,0.00126326,0.009860591],"category_scores_gemma":[0.003597108,0.000228997,0.0004686896,0.001336531,0.0005371794,0.0009375701,0.002075157,0.0009910421,0.01224016],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000647203,"about_ca_system_score_gemma":0.001347188,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01417201,"about_ca_topic_score_gemma":0.02045824,"domain_scores_codex":[0.9989457,0.0002721399,0.0001154799,0.0002201818,0.0002977506,0.000148816],"domain_scores_gemma":[0.9987638,0.00030293,0.00008518065,0.0003359569,0.0003772772,0.0001349065],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001604208,0.0005896406,0.01406216,0.002097376,0.0001691519,0.001173434,0.002146567,0.004460351,0.03091945,0.00573796,0.7018362,0.2352033],"study_design_scores_gemma":[0.0004040538,0.000368057,0.06939569,0.000395747,0.00009896402,0.001470752,0.002776546,0.01765266,0.02099598,0.004896762,0.8813249,0.0002198328],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1279017,0.0018282,0.02480881,0.001718137,0.00106422,0.001150168,0.8038745,0.01070992,0.0269444],"genre_scores_gemma":[0.08338336,0.0003835029,0.01790966,0.0002914302,0.0001537942,0.001704362,0.8874772,0.0004390543,0.008257512],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01417201,"threshold_uncertainty_score":0.032987,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1233076174698496,"score_gpt":0.2252155139685487,"score_spread":0.1019078964986991,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}