{"id":"W2014517133","doi":"10.1109/icassp.2014.6854820","title":"Regularized constrained maximum likelihood linear regression for speech recognition","year":2014,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Hidden Markov model; Embedding; Regularization (linguistics); Pattern recognition (psychology); Computer science; Feature vector; Feature (linguistics); Artificial intelligence; Graph; Speech recognition; Maximization; Mathematics; Mathematical optimization; Theoretical computer science","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":[0.0007598578,0.0001743603,0.0002342943,0.000147468,0.0001606181,0.0001252752,0.000411179,0.0001397698,0.0003489706],"category_scores_gemma":[0.0004238535,0.0001386202,0.0001574766,0.0002265074,0.00004540811,0.0003510353,0.00007438201,0.00009126399,0.0003935006],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002176117,"about_ca_system_score_gemma":0.00004816433,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007517506,"about_ca_topic_score_gemma":0.000007746629,"domain_scores_codex":[0.9985408,0.0001141258,0.000304017,0.0004596981,0.0002468193,0.000334532],"domain_scores_gemma":[0.9986365,0.0003806046,0.0001168995,0.000452413,0.0002550559,0.0001585122],"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":[0.00002390357,0.00005770131,0.000006847195,0.00001369131,0.00001060378,0.000002338328,0.00002880635,4.636572e-8,0.006265487,0.001971971,0.002269466,0.9893491],"study_design_scores_gemma":[0.003746677,0.0003941848,0.0001115064,0.000172498,0.00003524865,0.0001289634,0.00007038406,0.1810847,0.4889972,0.275141,0.04939687,0.0007207681],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007122225,0.000007179258,0.9579344,0.002904254,0.0004594968,0.0004195454,0.000009486299,0.0005192329,0.0306242],"genre_scores_gemma":[0.01832504,0.00001022085,0.978788,0.001436645,0.0002196204,0.00006045539,0.00003830679,0.00001732119,0.001104392],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9886284,"threshold_uncertainty_score":0.5652767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02884976113153596,"score_gpt":0.2597818766113098,"score_spread":0.2309321154797739,"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."}}