{"id":"W4410076795","doi":"10.2196/69431","title":"Collection and Analysis of Repeated Speech Samples: Methodological Framework and Example Protocol","year":2025,"lang":"en","type":"article","venue":"JMIR Research Protocols","topic":"Voice and Speech Disorders","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Wellcome Trust","keywords":"Preprint; Protocol (science); Computer science; Data collection; Natural language processing; Psychology; Data science; World Wide Web; Medicine; Statistics; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003251559,0.0001485762,0.0006380827,0.001145794,0.0002008464,0.00006503666,0.0001212665,0.0002841176,0.0002944177],"category_scores_gemma":[0.002735972,0.000113659,0.0001052199,0.004088347,0.0004005461,0.00007136526,0.0001905351,0.000662605,0.000002142705],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006404806,"about_ca_system_score_gemma":0.0002256418,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003110178,"about_ca_topic_score_gemma":0.00006117324,"domain_scores_codex":[0.9971769,0.0008964004,0.0004371704,0.0005330794,0.0005679405,0.0003884734],"domain_scores_gemma":[0.9973809,0.001391102,0.00008766251,0.0005059016,0.0004806943,0.0001537215],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.01425505,0.002874677,0.7943218,0.007713976,0.002732191,0.00009147595,0.001929069,0.0000058631,0.04093963,0.008248711,0.01645897,0.1104286],"study_design_scores_gemma":[0.009512286,0.006843235,0.6708235,0.005205194,0.0003630806,0.00002620645,0.001892261,0.00134808,0.04745902,0.02145252,0.2346275,0.0004471377],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.01799669,0.000005123888,0.002167442,0.0007673329,0.000001704797,0.9765347,0.000006680747,0.00005850717,0.00246186],"genre_scores_gemma":[0.001809724,0.000002592904,0.01794211,0.0001037311,0.00001858142,0.9787352,0.00001058908,0.00001123563,0.001366251],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.2181685,"threshold_uncertainty_score":0.463488,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.495173061314731,"score_gpt":0.6241466772327529,"score_spread":0.1289736159180218,"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."}}