{"id":"W2793646841","doi":"10.1109/icassp.2018.8462037","title":"Extracting Domain Invariant Features by Unsupervised Learning for Robust Automatic Speech Recognition","year":2018,"lang":"en","type":"preprint","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Autoencoder; Computer science; Robustness (evolution); Invariant (physics); Speech recognition; Word error rate; Artificial intelligence; Pattern recognition (psychology); Latent variable; Domain (mathematical analysis); Encoder; Natural language processing; Deep learning; Mathematics","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":["metaepi_narrow","scholarly_communication","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001556625,0.000472716,0.0005334698,0.000306585,0.0004188742,0.001094257,0.001141431,0.0005729147,0.0009495032],"category_scores_gemma":[0.0009026063,0.0004438683,0.0003159067,0.0002525226,0.00005704313,0.000483837,0.0006742286,0.0007981352,0.0002971345],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001372434,"about_ca_system_score_gemma":0.0001847543,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001101615,"about_ca_topic_score_gemma":0.00004038808,"domain_scores_codex":[0.996834,0.0003581736,0.0006297094,0.00114311,0.0004820282,0.0005529176],"domain_scores_gemma":[0.9973177,0.000952236,0.0004643178,0.0006973481,0.000378134,0.0001902335],"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":[0.0000140388,0.0001320872,0.00002036939,0.0002808768,0.0001146754,0.00001833696,0.0004399786,0.00002092427,0.001435592,0.0001443478,0.03204831,0.9653305],"study_design_scores_gemma":[0.002150183,0.0003722806,0.0005080554,0.001900696,0.00019715,0.0003667276,0.00109514,0.8225539,0.08149391,0.0761039,0.01025516,0.003002865],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01974351,0.0001161772,0.9566992,0.00156599,0.001201263,0.001257301,0.00003762005,0.00139043,0.01798852],"genre_scores_gemma":[0.005894803,0.00004738909,0.990518,0.0006262466,0.0004124984,0.0002939173,0.0003049901,0.00005900801,0.001843151],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9623276,"threshold_uncertainty_score":0.9999638,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05065405538082775,"score_gpt":0.2684047580639286,"score_spread":0.2177507026831008,"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."}}