{"id":"W6893545509","doi":"10.5281/zenodo.16729212","title":"MedTranscripts - A multimodal dataset of Spanish medical videos and time-aligned transcripts","year":2025,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Voice and Speech Disorders","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Pronunciation; Gold standard (test); Medical record; Human interaction","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008349703,0.0003230716,0.0006307378,0.0004685324,0.0006379814,0.000192109,0.001038122,0.0003913525,0.02481068],"category_scores_gemma":[0.00144188,0.000318986,0.0001152065,0.0004909444,0.0004271174,0.0001526008,0.0006410978,0.0005679555,0.0007690284],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006899638,"about_ca_system_score_gemma":0.00004554158,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002093678,"about_ca_topic_score_gemma":0.000005451659,"domain_scores_codex":[0.9968989,0.0003403604,0.0006121834,0.0006722654,0.001044479,0.0004317836],"domain_scores_gemma":[0.998026,0.00005368705,0.0001613631,0.0009080136,0.0004146008,0.0004363654],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003179371,0.0003937234,5.404565e-7,0.001222382,0.0002410277,0.0001242418,0.0001490775,3.645582e-7,0.001507688,0.00003575989,0.9846522,0.01135502],"study_design_scores_gemma":[0.00218929,0.0004993306,0.00004349642,0.000482124,0.0003210442,0.0002275087,0.0001019612,0.00009168653,0.0001871172,0.00001460162,0.9955959,0.0002459917],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0002859455,0.0003525461,0.0001589428,0.001647478,0.0001001622,0.0008403333,0.9926706,0.000195469,0.003748544],"genre_scores_gemma":[0.000516491,0.001077885,0.0001032907,0.0008765885,0.0001039966,1.047616e-7,0.9963985,0.0004472907,0.000475856],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.02404165,"threshold_uncertainty_score":0.9999262,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02059578377920814,"score_gpt":0.274566441788386,"score_spread":0.2539706580091779,"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."}}