{"id":"W3161950572","doi":"10.1109/icassp39728.2021.9413740","title":"TSTNN: Two-Stage Transformer Based Neural Network for Speech Enhancement in the Time Domain","year":2021,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":186,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Encoder; Computer science; Transformer; Speech recognition; Artificial neural network; Speech enhancement; Benchmark (surveying); Time domain; Pattern recognition (psychology); Artificial intelligence; Noise reduction; Computer vision; Engineering; Voltage","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":[0.0005779989,0.0001201368,0.0001344872,0.00002865584,0.0001349078,0.000235413,0.0005373267,0.00002957009,0.0002356803],"category_scores_gemma":[0.00001113747,0.00008212026,0.00008193795,0.0004956922,0.0000195313,0.0002489035,0.00002576769,0.00009859358,0.00003046026],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002412937,"about_ca_system_score_gemma":0.0001179896,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007603115,"about_ca_topic_score_gemma":0.00008285474,"domain_scores_codex":[0.9987531,0.00006795929,0.0002057667,0.000313786,0.0002375395,0.0004219067],"domain_scores_gemma":[0.9993989,0.0001560383,0.00003580055,0.0003225165,0.00004380641,0.00004290733],"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":[0.0001346078,0.0007968383,0.001010004,0.0001996666,0.00005650192,0.0005489076,0.002971113,0.01101428,0.1887357,0.01954081,0.02796655,0.747025],"study_design_scores_gemma":[0.002430344,0.0001653083,0.0002358289,0.00005759812,0.000008413822,0.00002636563,0.000159929,0.1567294,0.7881756,0.00779644,0.04380163,0.0004130534],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03905277,0.000167185,0.9394721,0.009769038,0.0001709569,0.0003042294,0.000001838928,0.00005630183,0.01100562],"genre_scores_gemma":[0.2316716,0.000005184185,0.7459075,0.01783003,0.0002842095,0.00009233312,0.00001826139,0.00001475239,0.004176093],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.746612,"threshold_uncertainty_score":0.3348767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01613232171435938,"score_gpt":0.2649973294156781,"score_spread":0.2488650077013187,"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."}}