{"id":"W2790603904","doi":"10.1016/j.neucom.2018.01.013","title":"Training and compensation of class-conditioned NMF bases for speech enhancement","year":2018,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Sherbrooke; McGill University","funders":"","keywords":"Computer science; Non-negative matrix factorization; Basis (linear algebra); Speech enhancement; Benchmark (surveying); Pattern recognition (psychology); Noise (video); Basis function; Residual; A priori and a posteriori; Artificial intelligence; Algorithm; Speech recognition; Matrix decomposition; Mathematics; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002395162,0.00009252892,0.0001439621,0.0001051241,0.0002265678,0.00009240473,0.0001986772,0.00002533204,0.000004942567],"category_scores_gemma":[0.00007445026,0.00009418106,0.00002988782,0.0002202949,0.00006135629,0.0002414008,0.00009621285,0.00005004849,0.00000226925],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001025512,"about_ca_system_score_gemma":0.00003946288,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001907524,"about_ca_topic_score_gemma":0.000001517349,"domain_scores_codex":[0.9990957,0.0000236261,0.0002350286,0.0002928412,0.0001441628,0.0002086927],"domain_scores_gemma":[0.9993278,0.0001917754,0.0001773099,0.0001403671,0.0001193453,0.00004345044],"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":[0.000006787542,0.00003073914,0.0002270649,0.00005752009,0.000007345774,0.000001918938,0.001157298,0.00002392019,0.1481149,0.001935958,0.0001433272,0.8482932],"study_design_scores_gemma":[0.0005495917,0.0003668412,0.001444582,0.0001040215,0.000006698175,0.00002344634,0.00007344757,0.08389485,0.9071698,0.002229156,0.003988105,0.0001495009],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4154764,0.00001710839,0.5826254,0.0002520935,0.000174544,0.0001397993,8.429046e-7,0.00006013899,0.001253674],"genre_scores_gemma":[0.7939942,0.000001004376,0.205253,0.0005413108,0.0001791385,0.000003847192,0.000001903769,0.00000579371,0.00001983559],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8481438,"threshold_uncertainty_score":0.3840592,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04543757921521153,"score_gpt":0.294166125607181,"score_spread":0.2487285463919695,"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."}}