{"id":"W2126853534","doi":"10.1109/icalip.2008.4590005","title":"Comparison of frame margin probability training to other discriminative methods for multi-condition speech recognition","year":2008,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique","funders":"","keywords":"Discriminative model; Margin (machine learning); Computer science; Speech recognition; Pattern recognition (psychology); Frame (networking); Interpolation (computer graphics); Artificial intelligence; Maximum likelihood; Linear interpolation; Machine learning; Mathematics; Statistics","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":[0.0008269956,0.0001453058,0.0003378862,0.0001704708,0.0001219978,0.0000293431,0.0002930083,0.00008673839,0.0002475862],"category_scores_gemma":[0.0006847616,0.0001288337,0.0001328981,0.0003059692,0.00007354288,0.0003593556,0.00005232786,0.00008756798,0.0000418723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005147176,"about_ca_system_score_gemma":0.0000467754,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003738745,"about_ca_topic_score_gemma":0.00003453657,"domain_scores_codex":[0.9985141,0.0002590304,0.0004122567,0.0004161835,0.000169415,0.0002289733],"domain_scores_gemma":[0.9986006,0.000584802,0.0001483068,0.0002701861,0.0002925695,0.000103504],"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":[0.00002467086,0.0003868317,0.0002377055,0.00002618807,0.00002493778,5.581059e-7,0.006828531,0.000003049169,0.008649482,0.002003562,0.000335182,0.9814793],"study_design_scores_gemma":[0.0009249273,0.000393813,0.0072664,0.00007682963,0.00002876541,0.00002102464,0.001453506,0.0964272,0.8640655,0.02682237,0.002085558,0.0004340793],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05584176,0.000008043332,0.9398418,0.0004085656,0.0001297593,0.0007944595,0.00002768963,0.0001269796,0.002820925],"genre_scores_gemma":[0.1277042,9.120284e-7,0.8715922,0.000306529,0.00002416197,0.0001520823,0.0000105231,0.00001016982,0.0001992369],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9810452,"threshold_uncertainty_score":0.5253686,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4009099348531255,"score_gpt":0.4637861807565178,"score_spread":0.0628762459033923,"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."}}