{"id":"W3205904637","doi":"10.1145/3474085.3475318","title":"Towards Realistic Visual Dubbing with Heterogeneous Sources","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph; Memorial University of Newfoundland","funders":"","keywords":"Computer science; Leverage (statistics); Speech recognition; Task (project management); Fidelity; Artificial intelligence; Flexibility (engineering); Natural language processing","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0001621374,0.0003204905,0.0003679974,0.0001097225,0.0001511556,0.001669378,0.001082632,0.0001669929,0.00005236319],"category_scores_gemma":[0.00003581355,0.0002503098,0.0001065827,0.0002350869,0.00005135524,0.0002041055,0.001897652,0.0004016585,0.0000119173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006453636,"about_ca_system_score_gemma":0.0005944283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002281552,"about_ca_topic_score_gemma":0.0000964654,"domain_scores_codex":[0.9978828,0.00005426461,0.000258509,0.0009114466,0.0004898038,0.0004031818],"domain_scores_gemma":[0.9987823,0.00003483477,0.0001755522,0.0006861457,0.000167906,0.0001532337],"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":[0.00005431687,0.0005436005,0.002556275,0.001972862,0.0007629541,0.005871185,0.007692527,0.05659011,0.005002357,0.001479475,0.0007079804,0.9167663],"study_design_scores_gemma":[0.0009216001,0.0004568738,0.001140559,0.002422021,0.0001445752,0.001733294,0.0005366221,0.1230546,0.8591117,0.006055512,0.00129828,0.003124339],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1861753,0.0007820576,0.806024,0.0004549452,0.0003682039,0.0001074306,0.000001400589,0.0004649996,0.005621686],"genre_scores_gemma":[0.7868906,0.00003096422,0.2119145,0.0005837504,0.0001800378,0.00001808208,0.00001593541,0.00002293899,0.0003432239],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.913642,"threshold_uncertainty_score":0.9999949,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01769955542518053,"score_gpt":0.264486470105342,"score_spread":0.2467869146801615,"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."}}