{"id":"W2036242736","doi":"10.1109/icassp.2013.6638952","title":"A deep convolutional neural network using heterogeneous pooling for trading acoustic invariance with phonetic confusion","year":2013,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":188,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Computer science; Word error rate; Speech recognition; Pooling; Convolutional neural network; TIMIT; Artificial intelligence; Dropout (neural networks); Deep learning; Artificial neural network; Hidden Markov model; Pattern recognition (psychology); Machine learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005888924,0.0009642357,0.0005129816,0.0003937139,0.0003447289,0.000567027,0.001341429,0.0007245412,0.002002529],"category_scores_gemma":[0.001025465,0.0004128813,0.0004938471,0.0005160319,0.0004572139,0.001221158,0.001066063,0.0008218795,0.0006409706],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009042357,"about_ca_system_score_gemma":0.001027798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006980081,"about_ca_topic_score_gemma":0.01408261,"domain_scores_codex":[0.9996825,0.00003155014,0.00001506482,0.0001045212,0.0001047176,0.0000616757],"domain_scores_gemma":[0.9997932,0.00005366368,0.00002400506,0.00004902461,0.00005773256,0.00002249186],"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":[0.0003174313,0.0002171709,0.002570908,0.0001549322,0.0002358794,0.0002906526,0.00008469886,0.3076307,0.1595135,0.01192551,0.006428211,0.5106303],"study_design_scores_gemma":[0.00001398822,0.00006746922,0.0005862334,0.000009980908,0.00003962398,0.00008659593,0.000005395121,0.9684488,0.02579026,0.002435148,0.002500912,0.00001561505],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03272922,0.0003933572,0.9619632,0.0001439036,0.00008299365,0.00005018734,0.0001883085,0.001829135,0.002619695],"genre_scores_gemma":[0.5793263,0.0003583436,0.4114891,0.0002608399,0.0000591615,0.0001364895,0.0008739488,0.0002111369,0.007284734],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006980081,"threshold_uncertainty_score":0.01387894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03207903847495538,"score_gpt":0.2313530530693191,"score_spread":0.1992740145943637,"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."}}