{"id":"W3124608786","doi":"10.1007/s00034-020-01604-6","title":"DNN-Based Calibrated-Filter Models for Speech Enhancement","year":2021,"lang":"en","type":"article","venue":"Circuits Systems and Signal Processing","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University; McGill University","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada","keywords":"Filter (signal processing); Speech enhancement; Wiener filter; Computer science; Intelligibility (philosophy); Spectrogram; Filter design; Speech recognition; Noise (video); Mathematics; Artificial intelligence; Computer vision","routes":{"ca_aff":true,"ca_fund":true,"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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.000374628,0.0002234946,0.0003198682,0.00008837508,0.0004240262,0.001391175,0.0003174424,0.0001031022,0.000007456405],"category_scores_gemma":[0.0000189518,0.0002051597,0.00006390935,0.0003686941,0.0000393295,0.0009982964,0.00007084074,0.0001154441,0.000003798272],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005118812,"about_ca_system_score_gemma":0.0005841241,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008452877,"about_ca_topic_score_gemma":0.000001811067,"domain_scores_codex":[0.9980969,0.00005261464,0.0003989984,0.0006673996,0.0003177202,0.0004663988],"domain_scores_gemma":[0.9989315,0.00008416166,0.0001794243,0.0002438362,0.0003928757,0.0001682385],"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":[0.000009532622,0.000152583,0.0002035998,0.002487599,0.00004075634,0.0001234113,0.0007962346,0.005208594,0.1836817,0.002610946,0.0006074471,0.8040776],"study_design_scores_gemma":[0.0006192989,0.00005898473,0.000009028117,0.0006649458,0.00001495196,0.000080488,0.0001133487,0.6624721,0.3315486,0.002947282,0.001124897,0.0003460802],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009655694,0.007463933,0.9802945,0.0002623149,0.0002110357,0.0002564937,0.000004499677,0.0001400154,0.001711468],"genre_scores_gemma":[0.9830523,0.000007711553,0.01553652,0.0005571406,0.0002059142,0.0000676071,0.000008600475,0.0000230171,0.0005411797],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9733966,"threshold_uncertainty_score":0.9996455,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03832709449995774,"score_gpt":0.2514362554248574,"score_spread":0.2131091609248996,"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."}}