{"id":"W4400427833","doi":"10.1080/2050571x.2024.2374160","title":"Leveraging natural language processing models to automate speech-intelligibility scoring","year":2024,"lang":"en","type":"article","venue":"Speech Language and Hearing","topic":"Hearing Loss and Rehabilitation","field":"Neuroscience","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Baycrest Hospital; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Computer science; Intelligibility (philosophy); Natural language processing; Speech recognition; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0005070911,0.0001975298,0.0002076482,0.0002224765,0.0002434913,0.000538688,0.0001429618,0.0000657765,0.00002622848],"category_scores_gemma":[0.0002133316,0.0001690481,0.00006717096,0.0004199762,0.00005706852,0.0005606132,0.0001763187,0.0003716382,0.00005810177],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001000078,"about_ca_system_score_gemma":0.00005458676,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006575736,"about_ca_topic_score_gemma":0.00002483262,"domain_scores_codex":[0.9982044,0.00005738692,0.0002631078,0.0007173638,0.0003000169,0.0004577968],"domain_scores_gemma":[0.9993861,0.0001595585,0.00002468068,0.0002535205,0.00002399607,0.0001521248],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000009255539,0.00001443756,0.0001300841,0.0004044951,0.000002838618,0.0002838463,0.02748632,0.0001762179,0.5512803,0.0002347628,0.00001267224,0.4199648],"study_design_scores_gemma":[0.000264006,0.0000906039,0.004598848,0.001664937,0.00002552471,0.0003722812,0.009536237,0.3707066,0.6102979,0.0014839,0.0002581555,0.0007009454],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9922103,0.002521452,0.001441098,0.0003556953,0.0003593955,0.0002937817,0.000003655422,0.0006503243,0.002164266],"genre_scores_gemma":[0.9939571,0.0000184426,0.004416844,0.000333814,0.0002226451,0.00001074984,0.0000016722,0.00003892484,0.0009998168],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4192638,"threshold_uncertainty_score":0.6893581,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04692329256596883,"score_gpt":0.3325447950866364,"score_spread":0.2856215025206675,"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."}}