{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-neural-parametric-singing-synthesizer","title":"A Neural Parametric Singing Synthesizer","arxiv_id":"1704.03809","date":"2017-04-12","proceeding":null,"authors":["Merlijn Blaauw","Jordi Bonada"],"abstract":"We present a new model for singing synthesis based on a modified version of\nthe WaveNet architecture. Instead of modeling raw waveform, we model features\nproduced by a parametric vocoder that separates the influence of pitch and\ntimbre. This allows conveniently modifying pitch to match any target melody,\nfacilitates training on more modest dataset sizes, and significantly reduces\ntraining and generation times. Our model makes frame-wise predictions using\nmixture density outputs rather than categorical outputs in order to reduce the\nrequired parameter count. As we found overfitting to be an issue with the\nrelatively small datasets used in our experiments, we propose a method to\nregularize the model and make the autoregressive generation process more robust\nto prediction errors. Using a simple multi-stream architecture, harmonic,\naperiodic and voiced/unvoiced components can all be predicted in a coherent\nmanner. We compare our method to existing parametric statistical and\nstate-of-the-art concatenative methods using quantitative metrics and a\nlistening test. While naive implementations of the autoregressive generation\nalgorithm tend to be inefficient, using a smart algorithm we can greatly speed\nup the process and obtain a system that's competitive in both speed and\nquality.","url_abs":"http://arxiv.org/abs/1704.03809v3","url_pdf":"http://arxiv.org/pdf/1704.03809v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-neural-parametric-singing-synthesizer","repo_url":"https://github.com/daksithj/vocsynth-neural-singing-synthesizer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"a-neural-parametric-singing-synthesizer","repo_url":"https://github.com/seaniezhao/torch_npss","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"dilated-causal-convolution","method_name":"Dilated Causal Convolution"},{"method_slug":"mixture-of-logistic-distributions","method_name":"Mixture of Logistic Distributions"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"wavenet","method_name":"WaveNet"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}