{"id":"W3206189675","doi":"10.1109/icassp43922.2022.9747814","title":"Large-Scale Self-Supervised Speech Representation Learning for Automatic Speaker Verification","year":2022,"lang":"en","type":"article","venue":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":102,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Generalizability theory; Computer science; Speech recognition; Word error rate; Speaker recognition; Artificial intelligence; Representation (politics); Artificial neural network; Scale (ratio); Feature (linguistics); Pattern recognition (psychology); Machine learning; Mathematics","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001268903,0.000387335,0.0004158143,0.000497583,0.001138796,0.0009149078,0.001139939,0.0001303504,0.002280774],"category_scores_gemma":[0.0002805972,0.0004215977,0.0001757487,0.0005865452,0.00008197835,0.0006909479,0.0003248943,0.0007746368,0.00007071548],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003448308,"about_ca_system_score_gemma":0.0003323498,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001935069,"about_ca_topic_score_gemma":0.000007875948,"domain_scores_codex":[0.9956914,0.0003185476,0.0007503577,0.001128938,0.001551735,0.0005590103],"domain_scores_gemma":[0.9977696,0.0004130426,0.0005150227,0.0004122565,0.0006603603,0.0002297243],"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":[0.0003298735,0.001527078,0.0009077172,0.0002445513,0.0002667115,0.0002017168,0.004040523,0.001639841,0.08344477,0.00520104,0.008806953,0.8933892],"study_design_scores_gemma":[0.00112663,0.0003428824,0.0004090668,0.00006620648,0.00006133707,0.0001789931,0.002469466,0.9826756,0.003558284,0.004618881,0.003977077,0.0005156475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1322812,0.0001250367,0.8490936,0.005613015,0.00192185,0.001210901,0.0002562636,0.0009461328,0.008551979],"genre_scores_gemma":[0.9063435,0.0001067145,0.08928773,0.0009742996,0.0003264602,0.0003628111,0.0002522927,0.00005593717,0.002290325],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9810357,"threshold_uncertainty_score":0.9998236,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0517667352431102,"score_gpt":0.3075118367600685,"score_spread":0.2557451015169583,"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."}}