{"id":"W4391831786","doi":"10.4995/eurocall2023.2023.17007","title":"Evaluating the effectiveness of Microsoft Transcribe for automating the assessment of pronunciation in language proficiency tests","year":2023,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Université du Québec à Montréal","keywords":"Computer science; Pronunciation; Natural language processing; Test (biology); Automation; Reliability (semiconductor); Artificial intelligence; Linguistics; Engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.01003718,0.000776027,0.0003566541,0.001293492,0.0004368532,0.001039752,0.0005666069,0.0007519113,0.001471019],"category_scores_gemma":[0.04092854,0.0002731638,0.0003614631,0.0004899433,0.0004852472,0.001010401,0.001008818,0.0003413471,0.0008086744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000413181,"about_ca_system_score_gemma":0.0007028831,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004663447,"about_ca_topic_score_gemma":0.01481189,"domain_scores_codex":[0.9909816,0.004876117,0.0009582391,0.000845086,0.00211802,0.0002209472],"domain_scores_gemma":[0.9609774,0.026469,0.00260694,0.001854513,0.007067108,0.001024988],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.01315559,0.003616774,0.276215,0.0009721927,0.0002952706,0.001020794,0.01649211,0.009175276,0.1546906,0.0008761437,0.002233136,0.5212572],"study_design_scores_gemma":[0.0007858482,0.03626252,0.7368425,0.0002641208,0.0006069744,0.002311934,0.009400018,0.05690725,0.1461574,0.0006448867,0.009458586,0.0003579573],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9892942,0.0001383076,0.008051844,0.00003864396,0.00002724102,0.0003003071,0.0001549086,0.0002428575,0.001751805],"genre_scores_gemma":[0.9590955,0.0001792152,0.03824222,0.00004734785,0.00002771593,0.0003013724,0.0003211447,0.0000987649,0.00168675],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01003718,"threshold_uncertainty_score":0.05308235,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0713846181357021,"score_gpt":0.4181764047821827,"score_spread":0.3467917866464806,"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."}}