{"id":"W2166688768","doi":"10.1109/icassp.2011.5947435","title":"Well-calibrated heavy tailed Bayesian speaker verification for microphone speech","year":2011,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure; Computer Research Institute of Montréal","funders":"","keywords":"NIST; Normalization (sociology); Computer science; Speech recognition; Microphone; Softmax function; Speaker recognition; Speaker verification; Bayesian probability; Feature extraction; Linear discriminant analysis; Classifier (UML); Phone; Discriminative model; Pattern recognition (psychology); Artificial intelligence; Artificial neural network","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":[0.0002022392,0.0001306596,0.0001447543,0.0001048475,0.00009240953,0.0001010493,0.0004675097,0.00008161159,0.002337527],"category_scores_gemma":[0.00003101528,0.000113677,0.00008857289,0.0003262905,0.00003105939,0.0003371348,0.00004652004,0.00005264887,0.001132997],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002208378,"about_ca_system_score_gemma":0.00004151177,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007813853,"about_ca_topic_score_gemma":0.00002949017,"domain_scores_codex":[0.9989547,0.00004652065,0.0002419474,0.0003805924,0.000129152,0.0002471574],"domain_scores_gemma":[0.9991806,0.00005526553,0.00006921547,0.0004715571,0.0001078334,0.0001155613],"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":[0.0003472686,0.001312758,0.00307186,0.0000770761,0.0001861479,0.00005297689,0.001891412,0.000001194995,0.1450803,0.04684807,0.08380757,0.7173234],"study_design_scores_gemma":[0.0006471306,0.00009616741,0.002824679,0.00001263587,0.00001228064,0.00003611525,0.0000772499,0.02342041,0.9445696,0.004830015,0.02313432,0.0003393363],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01261423,0.00001252626,0.93256,0.0007001202,0.0003514578,0.0003587677,0.000003642301,0.0003345362,0.05306477],"genre_scores_gemma":[0.2520779,0.00001225264,0.7423261,0.001079932,0.00005539995,0.00004789165,0.00001081911,0.00001831623,0.004371302],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7994893,"threshold_uncertainty_score":0.9996448,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04192580115728677,"score_gpt":0.2329112561070658,"score_spread":0.190985454949779,"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."}}