{"id":"W2810931900","doi":"10.1109/isspit.2017.8388664","title":"A reduced complexity MFCC-based deep neural network approach for speech enhancement","year":2017,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"PESQ; Computer science; Short-time Fourier transform; Mel-frequency cepstrum; Speech recognition; Speech enhancement; Benchmark (surveying); Artificial intelligence; Artificial neural network; Computational complexity theory; Pattern recognition (psychology); Feature (linguistics); Frequency domain; Cepstrum; Distortion (music); Feature extraction; Fourier transform; Noise reduction; Algorithm; Mathematics; Computer vision; Bandwidth (computing)","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":[0.0003557726,0.0001669611,0.0002030827,0.00002963071,0.000987066,0.0008133439,0.001590244,0.00005010229,0.00002225455],"category_scores_gemma":[0.00005473238,0.0001412639,0.00009878273,0.00008942997,0.000120143,0.0005737682,0.0002487196,0.00009401038,0.00001089549],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003470567,"about_ca_system_score_gemma":0.00006511941,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002461613,"about_ca_topic_score_gemma":0.0000199728,"domain_scores_codex":[0.9985105,0.00002189978,0.0002066445,0.0005032882,0.0002345076,0.0005231466],"domain_scores_gemma":[0.998549,0.00003274792,0.0001961253,0.0009980534,0.0001006956,0.0001233864],"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":[0.000121868,0.0005863191,0.00132732,0.0001762858,0.00007476014,0.00001751675,0.0002125209,0.005901068,0.02366011,0.01339804,0.01031197,0.9442122],"study_design_scores_gemma":[0.0007961936,0.0001054331,0.0007510908,0.00001145214,0.000007235277,0.000005673252,0.000006789573,0.8168466,0.1748807,0.005342435,0.0009841,0.0002622811],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01035176,0.00004982547,0.9757186,0.001918855,0.0004531496,0.0003624723,8.022054e-7,0.0001555808,0.01098892],"genre_scores_gemma":[0.409139,4.629432e-7,0.5895917,0.0006887291,0.0002229619,0.00003753185,0.000007166363,0.000006605684,0.0003058677],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9439499,"threshold_uncertainty_score":0.7843094,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07473564545120231,"score_gpt":0.3057028318327087,"score_spread":0.2309671863815064,"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."}}