{"id":"W1486315246","doi":"10.1109/isspit.2004.1433718","title":"Investigation into a mel subspace based front-end processing for robust speech recognition","year":2005,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique; Université de Moncton","funders":"","keywords":"Computer science; Speech recognition; Subspace topology; Speech enhancement; TIMIT; Noise (video); Front and back ends; Pattern recognition (psychology); Vocabulary; Noise reduction; Multilayer perceptron; Hidden Markov model; Set (abstract data type); Artificial intelligence; Speech processing; Artificial neural network","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":[0.0003972932,0.0001514089,0.0001341319,0.0001398224,0.0002688426,0.0004232135,0.0003402605,0.00008025185,0.00004663816],"category_scores_gemma":[0.000118307,0.0001401419,0.00005073409,0.0003131774,0.00004133059,0.001845157,0.00004281714,0.00009294728,0.00006894779],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009771269,"about_ca_system_score_gemma":0.0002455652,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003424039,"about_ca_topic_score_gemma":0.0003059048,"domain_scores_codex":[0.9987689,0.00002685978,0.0002268379,0.0004345817,0.0002471,0.0002957505],"domain_scores_gemma":[0.9992338,0.00006313328,0.0001346395,0.0002095514,0.0002334676,0.0001254179],"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":[0.000008668999,0.00002297927,0.0001546793,0.0000724923,0.000003100397,0.000001058612,0.0005125379,0.0002006956,0.01241814,0.00002975096,0.001825253,0.9847506],"study_design_scores_gemma":[0.0005281946,0.0000452241,0.0001008744,0.00009033639,0.000007825446,0.000005969013,0.00006542681,0.3349837,0.6581462,0.004571576,0.00123654,0.0002181391],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03196217,0.0001562188,0.9573242,0.008285896,0.00008907694,0.0002630901,0.000001132881,0.0003607852,0.001557424],"genre_scores_gemma":[0.1137014,0.000001874306,0.88344,0.002045406,0.0002124167,0.00004351301,0.00001932186,0.00001342912,0.0005226055],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9845325,"threshold_uncertainty_score":0.5714821,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04811698718316983,"score_gpt":0.2545845945459469,"score_spread":0.2064676073627771,"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."}}