{"id":"W4224933355","doi":"10.1109/icassp43922.2022.9746261","title":"Complex IRM-Aware Training for Voice Activity Detection Using Attention Model","year":2022,"lang":"en","type":"article","venue":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University; McGill University","funders":"","keywords":"Spectrogram; Computer science; Speech recognition; Voice activity detection; Noise (video); Artificial intelligence; Feature extraction; Pattern recognition (psychology); Artificial neural network; Speech processing; Image (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","sts"],"consensus_categories":[],"category_scores_codex":[0.0009347645,0.0004493737,0.000444503,0.0004819685,0.001751678,0.001019062,0.001138062,0.0001445962,0.0001811033],"category_scores_gemma":[0.0001144634,0.0005016364,0.0001742662,0.0005202805,0.0001429087,0.00108233,0.0003576298,0.0009436053,0.000006683073],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005399482,"about_ca_system_score_gemma":0.0005957894,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003564895,"about_ca_topic_score_gemma":0.00001443406,"domain_scores_codex":[0.9960324,0.0001492212,0.0005789135,0.001163454,0.001399792,0.0006762112],"domain_scores_gemma":[0.9980285,0.0001905972,0.0006209526,0.000318213,0.0006030074,0.0002387243],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002787491,0.0002623858,0.00004607083,0.0001071519,0.0000737485,0.00005068238,0.0005209558,0.02405879,0.6624286,0.000377532,0.0003020136,0.3114933],"study_design_scores_gemma":[0.0009047422,0.0003386628,0.0001436406,0.00009336213,0.00005495418,0.0001836041,0.0008258995,0.9778177,0.01279676,0.005984255,0.0002867622,0.00056966],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1351073,0.00005911162,0.8614187,0.001142345,0.0008176535,0.0003756659,0.0002136286,0.0002285095,0.0006370575],"genre_scores_gemma":[0.9586982,0.00001753472,0.0394821,0.0006302579,0.0003827774,0.0001237862,0.00006612911,0.00004935807,0.0005498324],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9537589,"threshold_uncertainty_score":0.9997435,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1525965693994567,"score_gpt":0.3427553212816414,"score_spread":0.1901587518821847,"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."}}