{"id":"W4299806750","doi":"10.1109/icme52920.2022.9859621","title":"Mel-Spectrogram Image-Based End-to-End Audio Deepfake Detection Under Channel-Mismatched Conditions","year":2022,"lang":"en","type":"article","venue":"2022 IEEE International Conference on Multimedia and Expo (ICME)","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"Computer Research Institute of Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Spectrogram; Codec; Robustness (evolution); Speech recognition; Spoofing attack; Speech coding; Jitter; Artificial intelligence; Computer network; Telecommunications","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000553678,0.001157639,0.0008170389,0.000960777,0.0003594495,0.0008008899,0.0009169198,0.001100713,0.003698501],"category_scores_gemma":[0.001782575,0.0001806353,0.0003163712,0.0002965262,0.0003302141,0.001069581,0.001088026,0.0008703853,0.002824427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003574017,"about_ca_system_score_gemma":0.0005914792,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001781408,"about_ca_topic_score_gemma":0.005318571,"domain_scores_codex":[0.9996212,0.0000403553,0.00001707988,0.00008127031,0.000139001,0.0001011719],"domain_scores_gemma":[0.9993306,0.000178223,0.00005736439,0.0001548359,0.0002241749,0.00005476244],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001850243,0.0004671155,0.003088062,0.0002161279,0.0001044113,0.0004426363,0.00008018145,0.03583173,0.161109,0.001178401,0.006105442,0.7895268],"study_design_scores_gemma":[0.00003768244,0.0004277403,0.004515827,0.00003623795,0.0000487131,0.0007809073,0.00008750715,0.7230294,0.26608,0.001657445,0.003259711,0.00003881111],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4214332,0.001092505,0.5521539,0.0003700488,0.0004695335,0.0002629349,0.001317984,0.01333728,0.009562761],"genre_scores_gemma":[0.8322571,0.0002745593,0.1559108,0.0001843251,0.00005770915,0.00007323575,0.001524612,0.0001465093,0.009571101],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003698501,"threshold_uncertainty_score":0.01237273,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03054497750112798,"score_gpt":0.2803384128655858,"score_spread":0.2497934353644578,"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."}}