{"id":"W4409190667","doi":"10.2139/ssrn.5206990","title":"Enhanced Audio-Visual Speech Processing with Posterior Sampling Methods in Recurrent Variational Autoencoders","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Carleton University","funders":"","keywords":"Audio visual; Speech recognition; Computer science; Sampling (signal processing); Speech processing; Artificial intelligence; Computer vision; Multimedia","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0007498997,0.0006138916,0.000592484,0.0002445677,0.0001984696,0.0007309734,0.000767307,0.0007812171,0.003229736],"category_scores_gemma":[0.002420093,0.0003936034,0.0007075975,0.0002970338,0.0003572878,0.0009383119,0.001000669,0.001182521,0.001033452],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002928062,"about_ca_system_score_gemma":0.0005302739,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003109313,"about_ca_topic_score_gemma":0.004563371,"domain_scores_codex":[0.9997003,0.00009803053,0.0000171407,0.00006186358,0.00009434222,0.00002833456],"domain_scores_gemma":[0.9993905,0.0003708963,0.00003143432,0.00005930957,0.0001218698,0.00002602351],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004450723,0.000131867,0.0006145086,0.0003206508,0.0001687275,0.0001567787,0.0002126042,0.4567828,0.08040704,0.03662771,0.003528918,0.4206033],"study_design_scores_gemma":[0.000004453212,0.00001267416,0.00008962165,0.000004816278,0.000006985352,0.00001470175,0.00000457762,0.9939884,0.003216062,0.002201962,0.0004516433,0.000004083801],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007451193,0.0002361665,0.9911906,0.00008683706,0.00005287962,0.00001241835,0.00004896548,0.0002099887,0.0007111131],"genre_scores_gemma":[0.4058017,0.0007397528,0.581894,0.000161376,0.0002348362,0.0001049146,0.0005496797,0.000378601,0.01013517],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003229736,"threshold_uncertainty_score":0.01080453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02052235375358828,"score_gpt":0.3559810023055207,"score_spread":0.3354586485519324,"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."}}