{"id":"W1885745269","doi":"10.1109/icassp.2005.1415064","title":"Discriminative Training Based on the Criterion of Least Phone Competing Tokens for Large Vocabulary Speech Recognition","year":2006,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University","keywords":"Security token; Computer science; Discriminative model; Hidden Markov model; Speech recognition; Word error rate; Phone; Normalization (sociology); Vocabulary; Sigmoid function; Artificial intelligence; Set (abstract data type); Generalization; Overfitting; Pattern recognition (psychology); Artificial neural network; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007436445,0.0001290057,0.0001703044,0.0001223706,0.0001935218,0.00007776915,0.0002748217,0.00006084017,0.0002521457],"category_scores_gemma":[0.0001770283,0.00009186284,0.0001262831,0.0002260157,0.00003541409,0.0002036737,0.0000369179,0.0001113624,0.00003137585],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000245708,"about_ca_system_score_gemma":0.00003848595,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003295639,"about_ca_topic_score_gemma":0.00004089213,"domain_scores_codex":[0.998839,0.0001333116,0.000272418,0.0002626647,0.000246798,0.0002457888],"domain_scores_gemma":[0.9985712,0.0008725596,0.0001264794,0.0002245558,0.0001727582,0.00003248393],"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":[0.0001158545,0.0008873405,0.0001466918,0.00009835733,0.00003266633,0.00001739698,0.002467846,0.00003068776,0.01957957,0.03299296,0.004287212,0.9393434],"study_design_scores_gemma":[0.001473259,0.0002840844,0.002435024,0.0003182015,0.00003054325,0.00001759238,0.003659684,0.5408755,0.4400564,0.008566158,0.001853527,0.0004300387],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08594742,0.000009176604,0.8634977,0.004474317,0.0002188375,0.0004702263,0.00007421121,0.0001437656,0.04516438],"genre_scores_gemma":[0.8736309,0.000001056068,0.1249274,0.00114804,0.00007801654,0.00004037739,0.00003890263,0.00001032162,0.0001249787],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9389134,"threshold_uncertainty_score":0.3746058,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06140262388372575,"score_gpt":0.26316123209178,"score_spread":0.2017586082080542,"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."}}