{"id":"W2532585045","doi":"10.1109/nlpke.2003.1275932","title":"Performance improvement of automatic speech recognition systems via multiple language models produced by sentence-based clustering","year":2004,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Cluster analysis; Vocabulary; Artificial intelligence; Speech recognition; Language model; Sentence; Natural language processing; Grammar; Self-organizing map","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":[0.0003252308,0.0001579079,0.0002100105,0.0001308728,0.00007070971,0.00007622023,0.0003142986,0.00005473932,0.00002897972],"category_scores_gemma":[0.00003333489,0.0001375946,0.00006427038,0.0002564754,0.00002396925,0.0005805666,0.00006583444,0.00006950391,0.00006196813],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001016987,"about_ca_system_score_gemma":0.00005770555,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004015417,"about_ca_topic_score_gemma":0.00002062196,"domain_scores_codex":[0.9985952,0.00003176273,0.0004008009,0.0003508015,0.0003688394,0.0002525546],"domain_scores_gemma":[0.9991516,0.0000517578,0.0001676791,0.0004116775,0.0001380316,0.00007922266],"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":[0.00001029845,0.000257866,0.00006350502,0.0003944192,0.00002337645,0.000006916402,0.0003390989,0.002380133,0.3026446,0.00001219027,0.0000528907,0.6938147],"study_design_scores_gemma":[0.0004481458,0.00007503828,0.00001926829,0.0001247552,0.000004499029,0.00000751458,0.00008583072,0.561894,0.4371932,0.00003481526,0.000001947267,0.0001109564],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5455413,0.00002077091,0.4530591,0.0001056321,0.0001574895,0.0003931375,0.000007115286,0.0002124756,0.0005029782],"genre_scores_gemma":[0.8961474,0.000006798679,0.1035406,0.0001288241,0.00002100736,0.00006611339,0.00001177856,0.00001095505,0.00006659456],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6937038,"threshold_uncertainty_score":0.5610944,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02378218736327504,"score_gpt":0.2199273004085662,"score_spread":0.1961451130452912,"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."}}