{"id":"W3120067823","doi":"10.1007/s00521-020-05672-2","title":"Deep neural network architectures for dysarthric speech analysis and recognition","year":2021,"lang":"en","type":"article","venue":"Neural Computing and Applications","topic":"Voice and Speech Disorders","field":"Medicine","cited_by":48,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Moncton","funders":"","keywords":"Speech recognition; Computer science; Dysarthria; Hidden Markov model; Intelligibility (philosophy); Convolutional neural network; Mel-frequency cepstrum; Mixture model; Artificial intelligence; Artificial neural network; Pattern recognition (psychology); Feature extraction","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":[0.00006843609,0.00009178225,0.0001909372,0.00006502449,0.0002460797,0.00003793076,0.00002583249,0.00004138955,0.000007955357],"category_scores_gemma":[0.00002946326,0.00008521454,0.00008149591,0.0005098823,0.00003983199,0.00001024922,0.00002950714,0.0001166772,0.000001414157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000004442649,"about_ca_system_score_gemma":0.000009853697,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001434397,"about_ca_topic_score_gemma":0.00008509255,"domain_scores_codex":[0.9993137,0.00002262105,0.0001491693,0.0002731723,0.0000688431,0.0001724572],"domain_scores_gemma":[0.999489,0.0001639179,0.00004677868,0.0001350923,0.00008219908,0.0000829953],"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":[0.00002418506,0.00005457763,0.02808068,0.00006947164,0.0001519609,0.000004438589,0.0001139133,0.003593516,0.0005480882,0.00006493954,0.0001197302,0.9671745],"study_design_scores_gemma":[0.001777064,0.0002247593,0.2191638,0.00003952236,0.002003551,0.0003434659,0.00041787,0.7651086,0.0009370993,0.006447274,0.003152444,0.0003845382],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9623568,0.001150216,0.03415332,0.00166351,0.00002169456,0.0003445811,0.000006278017,0.00006847884,0.0002350999],"genre_scores_gemma":[0.9888517,0.00003872595,0.009526432,0.001026967,0.0003015525,0.00003325596,0.000154485,0.000009909407,0.00005696817],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.96679,"threshold_uncertainty_score":0.3474948,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01927766120844045,"score_gpt":0.2914342240449873,"score_spread":0.2721565628365469,"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."}}