{"id":"W4402464101","doi":"10.11159/mvml24.115","title":"Development of a Multimodal Framework for Deepfake Detection: Combining Visual and Audio Analysis","year":2024,"lang":"en","type":"article","venue":"Proceedings of the World Congress on Electrical Engineering and Computer Systems and Science","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Audio visual; Audio analyzer; Speech recognition; Artificial intelligence; Human–computer interaction; Multimedia; Audio signal processing; Audio signal; Speech coding","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":[0.0004333752,0.0001229723,0.0002359174,0.0005023342,0.000134002,0.0004170286,0.0002744529,0.00003397174,3.776274e-8],"category_scores_gemma":[0.00006482432,0.00008966539,0.00004349095,0.002041687,0.0001094917,0.0002928052,0.0001763204,0.0001249932,6.081056e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000314449,"about_ca_system_score_gemma":0.00003191475,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003481202,"about_ca_topic_score_gemma":0.000001236303,"domain_scores_codex":[0.9988668,0.000002696769,0.0002515911,0.0003730302,0.0002998539,0.0002060175],"domain_scores_gemma":[0.999435,0.0001945918,0.00008517413,0.00007113243,0.0001259295,0.00008814999],"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":[0.00003071898,0.00006099008,0.0006262233,0.0007482024,0.0003087615,0.000001190064,0.0014417,0.002288169,0.01246531,0.3493455,0.00002058193,0.6326627],"study_design_scores_gemma":[0.00009013942,0.0001304589,0.00157031,0.0003251863,0.00002078821,0.00001412662,0.000007707625,0.985616,0.01166549,0.0001880401,0.0002647136,0.0001069862],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4545844,0.0004266805,0.5432295,0.0000543378,0.001388231,0.0002071929,5.373285e-7,0.00008670628,0.00002235559],"genre_scores_gemma":[0.9678754,0.00000493462,0.0320082,0.000008193998,0.00005483718,0.00002127204,4.099538e-8,0.000005008924,0.00002206651],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9833279,"threshold_uncertainty_score":0.4021416,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006963104688251368,"score_gpt":0.2269524833259458,"score_spread":0.2199893786376944,"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."}}