{"id":"W3161565382","doi":"10.1121/10.0004854","title":"On the compromise between noise reduction and speech/noise spatial information preservation in binaural speech enhancement","year":2021,"lang":"en","type":"article","venue":"The Journal of the Acoustical Society of America","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique; Université du Québec à Montréal","funders":"National Natural Science Foundation of China","keywords":"Noise reduction; Noise (video); Value noise; Binaural recording; Speech recognition; Subspace topology; Computer science; Reduction (mathematics); Gradient noise; Speech enhancement; Noise measurement; Spatial analysis; Colors of noise; Filter (signal processing); Algorithm; Mathematics; Noise floor; Artificial intelligence; Computer vision; Statistics","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":[0.0007875929,0.0001001482,0.0001885861,0.0000219192,0.0001922694,0.00008799169,0.0005834366,0.00004303136,0.00001120484],"category_scores_gemma":[0.000291548,0.00005058579,0.0001121217,0.0004099398,0.0002110239,0.0005546937,0.0002499938,0.0003861426,0.000002568014],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007413245,"about_ca_system_score_gemma":0.0001129819,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007309155,"about_ca_topic_score_gemma":7.98324e-7,"domain_scores_codex":[0.9985974,0.0001710335,0.0004322006,0.00008241575,0.0005505804,0.0001663487],"domain_scores_gemma":[0.9986123,0.000343885,0.0005112124,0.0002772656,0.0002094817,0.00004582195],"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":[0.0002253379,0.0003026648,0.0009542085,0.0001277116,0.0001557344,0.000003145193,0.00721825,0.009387188,0.1755415,0.00005583166,0.01109633,0.7949321],"study_design_scores_gemma":[0.0015726,0.000685217,0.02500395,0.0005232021,0.0001702419,0.0001631922,0.002969704,0.3919994,0.5620468,0.0136358,0.0009349707,0.0002949488],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4787724,0.00006931049,0.4985612,0.0222411,0.0001519846,0.0001303389,0.000001495443,0.00000604328,0.000066065],"genre_scores_gemma":[0.9477665,0.0001656722,0.05087724,0.001021088,0.0001357271,0.000001228508,8.043948e-7,0.000003797822,0.00002797767],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7946372,"threshold_uncertainty_score":0.2062829,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01486415115118895,"score_gpt":0.2462231559701087,"score_spread":0.2313590048189197,"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."}}