{"id":"W4399900412","doi":"10.18280/ria.380312","title":"Building a Corpus for the Underexplored Moroccan Dialect (CFMD) Through Audio Segmentations","year":2024,"lang":"en","type":"article","venue":"Revue d intelligence artificielle","topic":"Language, Linguistics, Cultural Analysis","field":"Arts and Humanities","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Linguistics; Computer science; Natural language processing; Speech recognition","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"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":[0.001465522,0.0008524598,0.0006360438,0.007178873,0.003191662,0.001568714,0.001013097,0.001307697,0.01294352],"category_scores_gemma":[0.005701731,0.0003140395,0.0003499335,0.003120154,0.001489814,0.001051708,0.00307581,0.0008602397,0.00515906],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0021111,"about_ca_system_score_gemma":0.002786367,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03788482,"about_ca_topic_score_gemma":0.07566946,"domain_scores_codex":[0.9983066,0.0003917611,0.0002014622,0.0005685345,0.0002839902,0.0002476059],"domain_scores_gemma":[0.9958898,0.00136134,0.0002450792,0.0005620542,0.001687624,0.0002541125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001262488,0.0005083455,0.04374596,0.006584562,0.00009038719,0.0167658,0.06103859,0.001329909,0.1403963,0.01467959,0.1648512,0.5487471],"study_design_scores_gemma":[0.0001256429,0.0001770338,0.1874692,0.001281645,0.0001189066,0.004302922,0.03457812,0.002714424,0.01703054,0.002564297,0.7494739,0.0001634828],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7345946,0.009415749,0.0379583,0.002328407,0.001811373,0.003478285,0.1408947,0.001776144,0.06774236],"genre_scores_gemma":[0.7233362,0.002058252,0.08683699,0.001034672,0.000716018,0.005939082,0.1608248,0.0008433898,0.01841056],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03788482,"threshold_uncertainty_score":0.07532865,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1090863902592334,"score_gpt":0.317473105228068,"score_spread":0.2083867149688347,"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."}}