{"id":"W1527831725","doi":"10.1109/sipnn.1994.344894","title":"A theory on optimal construction of dynamic features of speech for HMM-based speech recognition","year":2002,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Hidden Markov model; Speech recognition; Computer science; Weighting; Generalization; Domain (mathematical analysis); Dynamic programming; Artificial intelligence; Pattern recognition (psychology); Speech processing; Algorithm; Mathematics","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":[0.0003484432,0.000130487,0.0002185525,0.0002670437,0.00005015118,0.00002815897,0.0002687461,0.00009877636,0.0008130614],"category_scores_gemma":[0.0001901145,0.0001140755,0.0001639245,0.0002665034,0.0001001388,0.0001704389,0.00002096888,0.00007729695,0.0000397223],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002916393,"about_ca_system_score_gemma":0.00002270367,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008582158,"about_ca_topic_score_gemma":0.000008956503,"domain_scores_codex":[0.9989159,0.00008979651,0.0002936068,0.0002845532,0.0002494394,0.0001667336],"domain_scores_gemma":[0.9987178,0.0005164443,0.0001830042,0.0003055547,0.0002282724,0.00004894032],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006866815,0.0001303956,0.000006326456,0.00003550762,0.00002030085,0.000002681465,0.00003989587,0.00000421019,0.002475917,0.006482565,0.0004263849,0.9903072],"study_design_scores_gemma":[0.001292162,0.0006130879,0.0003775794,0.0001291079,0.00003525467,0.00008937157,0.0001639198,0.07165395,0.90325,0.02183325,0.0002832433,0.0002790671],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2009068,0.0000549794,0.7494415,0.001219786,0.0005671378,0.0008561755,0.0001077,0.0002620702,0.04658385],"genre_scores_gemma":[0.2963748,0.00001421075,0.7028529,0.0002013689,0.00002756325,0.00001983773,0.0000127067,0.00001010413,0.000486511],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9900281,"threshold_uncertainty_score":0.8902453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02699712628630202,"score_gpt":0.2480160064840671,"score_spread":0.2210188801977651,"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."}}