{"id":"W4390037801","doi":"10.1162/tacl_a_00627","title":"AfriSpeech-200: Pan-African Accented Speech Dataset for Clinical and General Domain ASR","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Benchmark (surveying); Computer science; Speech recognition; Domain (mathematical analysis); Set (abstract data type); Productivity; Natural language processing; Test set; Test (biology); Artificial intelligence; Biology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001612317,0.00285175,0.001160138,0.002197735,0.001039433,0.00117596,0.001630493,0.002290001,0.01418737],"category_scores_gemma":[0.004128498,0.0003860275,0.001045057,0.001272286,0.0006244834,0.001120482,0.001889749,0.001374727,0.02479672],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007158857,"about_ca_system_score_gemma":0.00152424,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01612459,"about_ca_topic_score_gemma":0.02227729,"domain_scores_codex":[0.9983625,0.0003709759,0.0002008902,0.0003778908,0.0004509769,0.0002367595],"domain_scores_gemma":[0.9982704,0.0004125146,0.0000833612,0.0004571973,0.0005982194,0.0001783551],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002729665,0.0009639089,0.006284671,0.001952268,0.0003156969,0.002089001,0.0004693149,0.007290827,0.03705096,0.0007965968,0.7566945,0.1833625],"study_design_scores_gemma":[0.002106995,0.002700012,0.1275389,0.001002795,0.000568519,0.01101543,0.003192896,0.08682817,0.0729942,0.003732404,0.6874828,0.0008367677],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1485341,0.003927663,0.01951498,0.001549177,0.00260002,0.001912362,0.7791665,0.02226699,0.02052819],"genre_scores_gemma":[0.05424718,0.0005261062,0.00978362,0.0002851176,0.0002632073,0.0009302304,0.9259215,0.0003962375,0.007646847],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01612459,"threshold_uncertainty_score":0.04746145,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05812305070298272,"score_gpt":0.3487872494558333,"score_spread":0.2906641987528506,"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."}}