{"id":"W4255900679","doi":"10.32920/ryerson.14664396","title":"Optimal generative and discriminative acoustic model training for speech recognition","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; University of Calgary","funders":"","keywords":"Hidden Markov model; Discriminative model; Computer science; Speech recognition; Generative model; Noise (video); Acoustic model; Estimator; Artificial intelligence; Pattern recognition (psychology); Speech processing; Mathematics; Generative grammar","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00115714,0.000547431,0.0006600787,0.0003395963,0.0002723033,0.0007058989,0.0007251788,0.0006564585,0.001367701],"category_scores_gemma":[0.002920214,0.0005148089,0.0007227947,0.0002574815,0.000550614,0.0009590531,0.0008990651,0.001293269,0.0004945318],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007851401,"about_ca_system_score_gemma":0.0008679185,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002050671,"about_ca_topic_score_gemma":0.003015695,"domain_scores_codex":[0.9994529,0.0002029631,0.0000246886,0.0001191745,0.0001563469,0.00004393327],"domain_scores_gemma":[0.9992651,0.0004707842,0.0000611847,0.00008462328,0.0000942375,0.00002404323],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004072397,0.00005423753,0.0002680512,0.0000511694,0.00001932899,0.0000234822,0.00008697503,0.8968067,0.006316346,0.02912197,0.0004632762,0.06674775],"study_design_scores_gemma":[9.05329e-7,0.000008325012,0.000028636,0.000001982118,0.000001554729,0.000003771097,0.000002411077,0.9974619,0.0005228739,0.001848612,0.0001172511,0.000001801795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01292582,0.0001034928,0.9857146,0.00007054224,0.000008910127,0.0000139101,0.00001741902,0.0001657623,0.0009795086],"genre_scores_gemma":[0.7112953,0.0003317698,0.2828113,0.0001146073,0.00003108518,0.0001416342,0.0002053648,0.0001357982,0.004933063],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002050671,"threshold_uncertainty_score":0.006119609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.115481456402756,"score_gpt":0.3142326164056291,"score_spread":0.1987511600028731,"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."}}