{"id":"W3082793320","doi":"10.1007/s11042-020-09563-5","title":"A hybrid speech enhancement system with DNN based speech reconstruction and Kalman filtering","year":2020,"lang":"en","type":"article","venue":"Multimedia Tools and Applications","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Concordia University","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Intelligibility (philosophy); Kalman filter; Speech enhancement; Speech recognition; Residual; Noise (video); Noise reduction; Hybrid system; Artificial intelligence; Pattern recognition (psychology); Machine learning; Algorithm","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":[],"consensus_categories":[],"category_scores_codex":[0.0000814801,0.000129755,0.0001448623,0.00003688328,0.0002097896,0.00027642,0.0001809261,0.00002630814,0.000009039889],"category_scores_gemma":[0.00001117236,0.0001124472,0.00001676569,0.0001781707,0.00005575272,0.0003193952,0.00007537068,0.00009113414,0.00001669107],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002439118,"about_ca_system_score_gemma":0.00004414293,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009446724,"about_ca_topic_score_gemma":0.000004554871,"domain_scores_codex":[0.9990639,0.00001380428,0.0001737144,0.0004299298,0.0001408364,0.0001778163],"domain_scores_gemma":[0.9994178,0.00005738617,0.00009114636,0.0002093988,0.00005104641,0.0001732201],"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":[0.000007079838,0.0000168559,0.0004391568,0.0001126527,0.000009009834,0.000007225668,0.0001065442,0.00001378581,0.01589231,0.0001620583,0.00005828332,0.983175],"study_design_scores_gemma":[0.001317964,0.0001738113,0.001397151,0.0002304927,0.00003084126,0.0002229461,0.0002552207,0.3119306,0.6730886,0.0001166776,0.01072775,0.0005079634],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08245943,0.0001156473,0.9146866,0.00144974,0.00003212333,0.0004595911,0.00001464711,0.0001856147,0.0005966011],"genre_scores_gemma":[0.5721927,0.00001992218,0.4271481,0.0002821322,0.0001379756,0.0001809986,0.00001282113,0.000008859819,0.00001656191],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9826671,"threshold_uncertainty_score":0.4585463,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02015041982085012,"score_gpt":0.2236489106702159,"score_spread":0.2034984908493657,"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."}}