{"id":"W4389518357","doi":"10.18653/v1/2023.arabicnlp-1.38","title":"VoxArabica: A Robust Dialect-Aware Arabic Speech Recognition System","year":2023,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Alliance de recherche numérique du Canada; Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Arabic; Computer science; Upload; Natural language processing; Interface (matter); Speech recognition; Range (aeronautics); Artificial intelligence; Language model; Linguistics; World Wide Web; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0006318414,0.001410608,0.001072849,0.001083242,0.0005049978,0.001003122,0.001191079,0.0008769321,0.009770142],"category_scores_gemma":[0.001491684,0.0003858921,0.0005636254,0.0003058623,0.0002883655,0.00125936,0.001756186,0.00103614,0.01672905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003725638,"about_ca_system_score_gemma":0.0005625075,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002723971,"about_ca_topic_score_gemma":0.00246323,"domain_scores_codex":[0.9994898,0.0000720775,0.00004722398,0.0002272714,0.0001144601,0.00004913089],"domain_scores_gemma":[0.9995635,0.00006487741,0.00003150135,0.00009845894,0.0001947867,0.00004687563],"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":[0.001567228,0.000323327,0.004277132,0.0006499696,0.0002842823,0.001263276,0.0007711884,0.01115562,0.1850994,0.002213327,0.1313323,0.661063],"study_design_scores_gemma":[0.0005756994,0.0009470683,0.01226583,0.0002288592,0.0003389239,0.0027792,0.0008424066,0.565635,0.2097356,0.008811167,0.1973202,0.0005201375],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"software","genre_gemma":"empirical","genre_scores_codex":[0.1133079,0.003064699,0.3378733,0.000634891,0.000956799,0.0007627191,0.01615839,0.5072437,0.01999766],"genre_scores_gemma":[0.536824,0.0007622079,0.3635902,0.00152984,0.0003801908,0.0008514834,0.05655015,0.005111152,0.03440071],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009770142,"threshold_uncertainty_score":0.03268445,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05897081501115145,"score_gpt":0.234406232164648,"score_spread":0.1754354171534966,"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."}}