{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001236318,0.001007057,0.0009044678,0.0007023262,0.001213271,0.001104904,0.007758209,0.0006442005,0.0007862388],"category_scores_gemma":[0.0004307189,0.0009418603,0.0007108137,0.00234375,0.0007580668,0.0009148462,0.005958196,0.001720649,0.001607323],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003140852,"about_ca_system_score_gemma":0.001070235,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00127427,"about_ca_topic_score_gemma":0.0005647614,"domain_scores_codex":[0.9936346,0.0009372605,0.0007320972,0.003192197,0.0003979706,0.00110589],"domain_scores_gemma":[0.9926956,0.001196965,0.0008930751,0.004227741,0.0004957925,0.0004908295],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001544917,0.003695842,0.005697875,0.001406965,0.00477891,0.009484075,0.02425897,0.05510479,0.0001656352,0.07652946,0.03114535,0.7861872],"study_design_scores_gemma":[0.001958441,0.0001598947,0.001289032,0.0006345088,0.001153954,0.00005354341,0.009644696,0.8400956,0.00252093,0.01010851,0.1298466,0.002534269],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.46636,0.0007508668,0.3957594,0.003614431,0.01184035,0.005242925,0.02400301,0.001534227,0.0908948],"genre_scores_gemma":[0.9834367,0.0005357024,0.004044374,0.0008535399,0.000340699,0.000003707181,0.0005801112,0.00003203046,0.01017313],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7849908,"threshold_uncertainty_score":0.9999321,"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."}}