{"id":"W1902248030","doi":"10.1109/icassp.1995.479599","title":"Use of generalized dynamic feature parameters for speech recognition: maximum likelihood and minimum classification error approaches","year":2002,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Weighting; Speech recognition; Preprocessor; TIMIT; Hidden Markov model; Pattern recognition (psychology); Speech processing; Mel-frequency cepstrum; Word error rate; Feature (linguistics); Artificial intelligence; Feature extraction","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":[0.0001827188,0.0001692721,0.0002347815,0.0001691047,0.00008492349,0.0001633987,0.0002445826,0.0001436829,0.00008356423],"category_scores_gemma":[0.0001180998,0.000148325,0.0001174922,0.0002635214,0.00006861149,0.0005506505,0.00004722604,0.00007845648,0.00002629276],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002436676,"about_ca_system_score_gemma":0.0000122855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008001537,"about_ca_topic_score_gemma":0.00002037135,"domain_scores_codex":[0.998772,0.00007802117,0.0002690698,0.0004557224,0.0001880665,0.0002370658],"domain_scores_gemma":[0.9989715,0.0002725685,0.00014911,0.0003758708,0.0001244116,0.0001065914],"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":[0.00001862145,0.000138086,0.00005344719,0.00003845329,0.00003831699,0.000001726001,0.0001599626,5.26354e-7,0.002210307,0.0003788774,0.003895202,0.9930665],"study_design_scores_gemma":[0.001318861,0.0002043514,0.001446445,0.00005081922,0.00007216533,0.00009361342,0.0002574649,0.9459731,0.03125152,0.01505765,0.003739052,0.0005349456],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3705056,0.000300938,0.6089223,0.01467512,0.0004528915,0.001589332,0.0001126149,0.0003908023,0.003050429],"genre_scores_gemma":[0.09660368,0.0001081842,0.9016116,0.0004320966,0.00001982031,0.00009503383,0.00003515515,0.00001534741,0.001079105],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9925315,"threshold_uncertainty_score":0.6048518,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1978869977534294,"score_gpt":0.2676775401728769,"score_spread":0.06979054241944743,"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."}}