{"id":"W2123798005","doi":"10.1109/icassp.2010.5495646","title":"Multilingual acoustic modeling for speech recognition based on subspace Gaussian Mixture Models","year":2010,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":160,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Subspace topology; Computer science; Phone; Gaussian; Mixture model; Speech recognition; Set (abstract data type); Language model; Artificial intelligence; Acoustic model; Space (punctuation); Hidden Markov model; Natural language processing; Pattern recognition (psychology); Speech processing; Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004127781,0.0002080482,0.0001814518,0.0002107022,0.0001810037,0.0001941427,0.0004431274,0.0001964431,0.0001600793],"category_scores_gemma":[0.0002554482,0.0001800428,0.0001471685,0.0002083339,0.00002111371,0.0004200199,0.0000340876,0.0002804894,0.0001021577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002291958,"about_ca_system_score_gemma":0.00009164342,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002248266,"about_ca_topic_score_gemma":0.000189249,"domain_scores_codex":[0.9985311,0.00003705748,0.0002331747,0.0005422023,0.0003035287,0.0003529551],"domain_scores_gemma":[0.9986911,0.0003346988,0.00006308531,0.0004938963,0.0002488419,0.000168398],"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":[0.00007507773,0.0002866925,0.000004544442,0.00003629785,0.00001522996,0.00001702024,0.0001896961,0.01951184,0.02080236,0.001307637,0.00060509,0.9571485],"study_design_scores_gemma":[0.0005239919,0.00005966036,0.000002340027,0.00002670211,0.00001177944,0.00001113875,0.00003978732,0.9110518,0.07773715,0.0101746,0.0001086077,0.0002524492],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04771985,0.000002774935,0.9412403,0.001162444,0.0005957349,0.0005251132,0.00001948675,0.0003817787,0.008352594],"genre_scores_gemma":[0.5557336,0.000001408834,0.442945,0.0008885824,0.0001172392,0.0000599417,0.00001485087,0.00001592248,0.0002235452],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9568961,"threshold_uncertainty_score":0.734193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04797919133278814,"score_gpt":0.2764242881821557,"score_spread":0.2284450968493676,"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."}}