{"id":"W2898963093","doi":"10.1109/icassp.2019.8683800","title":"End-to-end Sound Source Separation Conditioned on Instrument Labels","year":2019,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Monaural; Source separation; Separation (statistics); Bottleneck; Computer science; End-to-end principle; Artificial intelligence; Speech recognition; Acoustics; Machine learning; Physics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001273123,0.0000931981,0.00009163644,0.00008410704,0.00009983304,0.0002667106,0.0002893808,0.00003304156,0.0004919967],"category_scores_gemma":[0.00001128486,0.00007994629,0.0000268597,0.0002187254,0.00001178763,0.0004399331,0.00008326732,0.00006352039,0.002691527],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004909393,"about_ca_system_score_gemma":0.00005541442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001147035,"about_ca_topic_score_gemma":0.000008834174,"domain_scores_codex":[0.9990904,0.00001964612,0.0001325885,0.0002928364,0.0002699125,0.0001945606],"domain_scores_gemma":[0.9994974,0.00004175423,0.00004966818,0.0002875222,0.00003996077,0.00008369391],"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":[0.00007498843,0.0005160351,0.01213786,0.00008452804,0.0001009002,0.00001760041,0.004381477,0.00671523,0.2214732,0.1313705,0.01996051,0.6031672],"study_design_scores_gemma":[0.0009850124,0.0005094516,0.004031542,0.00005140679,0.000004882147,0.00001683664,0.000100897,0.007059705,0.9440707,0.01074407,0.03199403,0.0004314486],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6387983,0.000006938126,0.3344835,0.001755184,0.0002279593,0.0001494396,9.711504e-7,0.0001535169,0.02442416],"genre_scores_gemma":[0.9748138,5.043293e-7,0.01592887,0.003427983,0.00004660372,0.000007861192,0.000005053794,0.000005152803,0.005764135],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7225975,"threshold_uncertainty_score":0.998085,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01678665798965863,"score_gpt":0.2705405133659615,"score_spread":0.2537538553763029,"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."}}