{"id":"W4297841867","doi":"10.21437/interspeech.2022-10761","title":"Daft-Exprt: Cross-Speaker Prosody Transfer on Any Text for Expressive Speech Synthesis","year":2022,"lang":"en","type":"article","venue":"Interspeech 2022","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ubisoft (Canada)","funders":"","keywords":"Prosody; Computer science; Speech synthesis; Speech recognition; Transfer (computing); 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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008573191,0.0003929927,0.0004294277,0.0004246211,0.0006859229,0.0004187241,0.002049583,0.0001042277,0.005776633],"category_scores_gemma":[0.0003410321,0.0003905981,0.0004714908,0.0006280034,0.000108084,0.0004684788,0.0004892619,0.0004905102,0.0003491863],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003428696,"about_ca_system_score_gemma":0.0001254577,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002306814,"about_ca_topic_score_gemma":0.000007449617,"domain_scores_codex":[0.9963897,0.0002781676,0.0005293026,0.001169008,0.0009239573,0.0007099057],"domain_scores_gemma":[0.9974507,0.0009381259,0.0001158643,0.001082437,0.0002064688,0.0002064206],"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":[0.001265865,0.002646917,0.0003986921,0.0001432506,0.000507167,0.0008311617,0.007060428,0.000175772,0.06669888,0.02606915,0.1162676,0.7779351],"study_design_scores_gemma":[0.001826317,0.001090686,0.0005410649,0.0001085789,0.00005752286,0.0003311233,0.002407758,0.006869832,0.7281638,0.003855851,0.2534267,0.001320777],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5953729,0.0004750758,0.2971759,0.01531893,0.007928592,0.00591102,0.001752002,0.002293437,0.07377215],"genre_scores_gemma":[0.9598255,0.00001987254,0.02409899,0.003436728,0.0002958142,0.003195804,0.00003248989,0.0001061262,0.008988721],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7766144,"threshold_uncertainty_score":0.9998546,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02503636111819494,"score_gpt":0.2806554597688282,"score_spread":0.2556190986506332,"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."}}