{"id":"W4229439535","doi":"10.18280/ts.390226","title":"Speech Enhancement for Robust Speech Recognition Using Weighted Low Rank and Sparse Decomposition Models under Low SNR Conditions","year":2022,"lang":"en","type":"article","venue":"Traitement du signal","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Noise (video); Speech enhancement; Computer science; Matrix decomposition; Speech recognition; Rank (graph theory); Low-rank approximation; Pattern recognition (psychology); Noise measurement; Robust principal component analysis; Word error rate; Artificial intelligence; Noise reduction; Mathematics; Principal component analysis; Hankel matrix; Eigenvalues and eigenvectors","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004713773,0.0002302345,0.0002222881,0.0002195062,0.0009322257,0.0002696713,0.0003043083,0.00004720559,0.0005088436],"category_scores_gemma":[0.000003411205,0.0002546344,0.000091335,0.0003505689,0.00004916379,0.000960189,0.0001582795,0.000159874,0.000007652665],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002409648,"about_ca_system_score_gemma":0.0001192209,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001422165,"about_ca_topic_score_gemma":0.000004984862,"domain_scores_codex":[0.9980043,0.00009731424,0.0004086493,0.0005706191,0.0004887709,0.0004304059],"domain_scores_gemma":[0.9992512,0.0000857789,0.0002018799,0.0001902455,0.0001439995,0.0001268314],"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":[0.0004113694,0.002225996,0.00008947191,0.0003837579,0.0003021062,0.0001110349,0.001400678,0.109025,0.5449597,0.002126739,0.002526699,0.3364375],"study_design_scores_gemma":[0.002204683,0.0003152652,0.00005919158,0.0001118834,0.00006137257,0.0001167803,0.0001923852,0.7408942,0.2290497,0.02649122,0.00008899507,0.0004143867],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3767259,0.00006662676,0.6218935,0.00039096,0.0001672036,0.0005108153,0.00005755233,0.00007695692,0.0001104356],"genre_scores_gemma":[0.7006397,0.00001375852,0.2979085,0.0008588748,0.0001355038,0.0001677133,0.0002205864,0.00002059111,0.00003481906],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6318692,"threshold_uncertainty_score":0.9999906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05036844168348995,"score_gpt":0.2753111501342437,"score_spread":0.2249427084507538,"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."}}