{"id":"W4408354038","doi":"10.1109/icassp49660.2025.10889723","title":"Data-Driven White Noise Gain Constrained Robust Superdirective Beamformer for Speech Enhancement","year":2025,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique; Université du Québec à Montréal","funders":"National Key Research and Development Program of China; Shenzhen Research Institute of Big Data","keywords":"Speech enhancement; White noise; Computer science; Noise (video); Speech recognition; Acoustics; Noise reduction; Artificial intelligence; Physics; Telecommunications","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":[0.0003364894,0.0001690785,0.000212053,0.0001254568,0.0002040892,0.0002515392,0.001245293,0.00005531447,0.00006741223],"category_scores_gemma":[0.00008975972,0.0001450787,0.00005525717,0.0004245024,0.00007533476,0.0009897182,0.0005048379,0.00008924474,0.00002739245],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007324645,"about_ca_system_score_gemma":0.0002873452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001189124,"about_ca_topic_score_gemma":0.00007290506,"domain_scores_codex":[0.998502,0.00001981658,0.0002583552,0.0006453786,0.0001838424,0.0003906545],"domain_scores_gemma":[0.9988395,0.00009390706,0.00006036611,0.000768109,0.0001620443,0.00007606118],"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":[0.00007697919,0.0004803677,0.001824403,0.000240012,0.0003306996,0.00001717901,0.0008141031,0.00127889,0.06299805,0.01785093,0.09359054,0.8204979],"study_design_scores_gemma":[0.002253421,0.000193206,0.0003529465,0.0002206319,0.0000603136,0.00001667521,0.000423541,0.2120084,0.70964,0.006329169,0.06776282,0.0007389102],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001672131,0.00007363282,0.9626606,0.00396039,0.0004338824,0.0004091882,0.00002996492,0.0001734069,0.03058678],"genre_scores_gemma":[0.03892832,0.00001255858,0.9483073,0.002017098,0.00008400042,0.00005565085,0.00007118392,0.000008769653,0.01051506],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.819759,"threshold_uncertainty_score":0.5916136,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03401116000292457,"score_gpt":0.2923598451668958,"score_spread":0.2583486851639712,"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."}}