{"id":"W2114948162","doi":"10.1109/icassp.1996.540277","title":"HMM-based speech recognition using state-dependent, linear transforms on Mel-warped DFT features","year":2002,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"TIMIT; Hidden Markov model; Computer science; Speech recognition; Word error rate; Pattern recognition (psychology); Mel-frequency cepstrum; Transformation (genetics); Feature extraction; Artificial intelligence; Feature (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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0002953275,0.000253728,0.0002324878,0.0003125738,0.0001934185,0.000218522,0.0004496987,0.0001283291,0.002331793],"category_scores_gemma":[0.00006415854,0.0002070134,0.0001638807,0.0004325768,0.00003684878,0.000533484,0.00002544755,0.0002368014,0.001181257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007685117,"about_ca_system_score_gemma":0.00003278734,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001004405,"about_ca_topic_score_gemma":0.0001063874,"domain_scores_codex":[0.9980746,0.0001053801,0.0003169838,0.0005300429,0.0005615733,0.0004114007],"domain_scores_gemma":[0.9990091,0.0001736035,0.00008534415,0.0004205191,0.0001339506,0.0001774619],"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":[0.00003021924,0.0002965444,0.00002147705,0.00001664071,0.00002585226,0.00008120115,0.0001917247,0.0001066273,0.002218793,0.00007435261,0.001219706,0.9957169],"study_design_scores_gemma":[0.001574123,0.0002545177,0.0001499249,0.00009070971,0.00002548507,0.00008887712,0.00007059239,0.3432706,0.6498495,0.001938473,0.002019997,0.0006671104],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1874633,0.00005904939,0.7414941,0.002712709,0.000878686,0.0007803859,0.00006149427,0.001367653,0.06518262],"genre_scores_gemma":[0.4553038,0.00009463482,0.534886,0.005479519,0.0002194774,0.00002947026,0.0000314275,0.00006427692,0.003891366],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9950498,"threshold_uncertainty_score":0.9995964,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0734056754320572,"score_gpt":0.2641613885171936,"score_spread":0.1907557130851364,"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."}}