{"id":"W2627724237","doi":"10.1109/icassp.2017.7952567","title":"Single-channel enhancement of convolutive noisy speech based on a discriminative NMF algorithm","year":2017,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke; McGill University","funders":"","keywords":"Non-negative matrix factorization; Discriminative model; Computer science; Expectation–maximization algorithm; Pattern recognition (psychology); Artificial intelligence; Algorithm; Speech enhancement; Benchmark (surveying); Channel (broadcasting); Matrix decomposition; Speech recognition; Mathematics; Maximum likelihood; Eigenvalues and eigenvectors; Noise reduction","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00116361,0.0009151248,0.001048091,0.000497041,0.0003007327,0.0003832249,0.0008712798,0.0009424799,0.001259831],"category_scores_gemma":[0.001943621,0.0003658496,0.0007983347,0.0004437114,0.0006043821,0.0009819368,0.000670285,0.0008972983,0.0007972696],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000328777,"about_ca_system_score_gemma":0.0005124843,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001076173,"about_ca_topic_score_gemma":0.001840994,"domain_scores_codex":[0.9994729,0.0001314163,0.00002362266,0.0001410214,0.0001879216,0.0000432075],"domain_scores_gemma":[0.9993408,0.0002900144,0.00005407211,0.000104994,0.0001788396,0.0000312369],"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":[0.0004224707,0.000183052,0.0007663908,0.0002182319,0.00009897767,0.0001689335,0.0001367496,0.2139589,0.09310552,0.0103775,0.002598088,0.6779652],"study_design_scores_gemma":[0.000009552985,0.00004922042,0.0002644279,0.000007310306,0.00001378761,0.0001222807,0.000006958626,0.9833084,0.01318877,0.001533752,0.001484308,0.00001122082],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004975996,0.0001904745,0.9940099,0.00004170576,0.00003740141,0.00001612399,0.00001465552,0.0002331299,0.0004805915],"genre_scores_gemma":[0.1623878,0.0003317037,0.8339162,0.0001504923,0.00007733541,0.00009611771,0.0001662381,0.0001364704,0.002737738],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001259831,"threshold_uncertainty_score":0.006153882,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03491364649410964,"score_gpt":0.2769887975592842,"score_spread":0.2420751510651745,"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."}}