{"id":"W4409450725","doi":"10.1016/j.specom.2025.103230","title":"Neural Chinese silent speech recognition with facial electromyography","year":2025,"lang":"en","type":"article","venue":"Speech Communication","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Natural Science Foundation of China","keywords":"Speech recognition; Electromyography; Facial electromyography; Computer science; Hidden Markov model; Artificial intelligence; Pattern recognition (psychology); Physical medicine and rehabilitation; Facial expression; Medicine","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":[0.0004038709,0.0002143604,0.0002148788,0.0004293406,0.0003296907,0.0002583475,0.001270507,0.0001010063,0.00006791648],"category_scores_gemma":[0.0001012319,0.0001829169,0.0001139452,0.001555812,0.00008704716,0.0006611855,0.0002290922,0.0003189502,0.0001028256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007505637,"about_ca_system_score_gemma":0.00008019657,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001014024,"about_ca_topic_score_gemma":0.0002243555,"domain_scores_codex":[0.9984303,0.000273595,0.0003065896,0.0003755894,0.0003245982,0.0002893437],"domain_scores_gemma":[0.9979506,0.0002082843,0.0001393809,0.001343247,0.0002863035,0.00007215618],"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":[0.00003648844,0.0001552008,0.001649575,0.00001168661,0.00004256879,0.000005667672,0.0001105106,0.000001479966,0.001389112,0.001203715,0.0006909366,0.9947031],"study_design_scores_gemma":[0.008372426,0.00142435,0.1829309,0.001043789,0.0003223887,0.0006703866,0.0007178404,0.06370573,0.4033119,0.2775002,0.05610684,0.003893213],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8164528,0.0006767344,0.06585564,0.01044565,0.0004079105,0.001049316,0.00001607788,0.001139685,0.1039562],"genre_scores_gemma":[0.8014934,0.0002654075,0.1966169,0.001056322,0.00003638656,0.00009285045,0.00008781493,0.00001412539,0.0003367763],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9908099,"threshold_uncertainty_score":0.7459134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01252535883379708,"score_gpt":0.2535649671924412,"score_spread":0.2410396083586441,"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."}}