{"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/universal-melgan-a-robust-neural-vocoder-for","title":"Universal MelGAN: A Robust Neural Vocoder for High-Fidelity Waveform Generation in Multiple Domains","arxiv_id":"2011.09631","date":"2020-11-19","proceeding":null,"authors":["Won Jang","Dan Lim","Jaesam Yoon"],"abstract":"We propose Universal MelGAN, a vocoder that synthesizes high-fidelity speech in multiple domains. To preserve sound quality when the MelGAN-based structure is trained with a dataset of hundreds of speakers, we added multi-resolution spectrogram discriminators to sharpen the spectral resolution of the generated waveforms. This enables the model to generate realistic waveforms of multi-speakers, by alleviating the over-smoothing problem in the high frequency band of the large footprint model. Our structure generates signals close to ground-truth data without reducing the inference speed, by discriminating the waveform and spectrogram during training. The model achieved the best mean opinion score (MOS) in most scenarios using ground-truth mel-spectrogram as an input. Especially, it showed superior performance in unseen domains with regard of speaker, emotion, and language. Moreover, in a multi-speaker text-to-speech scenario using mel-spectrogram generated by a transformer model, it synthesized high-fidelity speech of 4.22 MOS. These results, achieved without external domain information, highlight the potential of the proposed model as a universal vocoder.","url_abs":"https://arxiv.org/abs/2011.09631v2","url_pdf":"https://arxiv.org/pdf/2011.09631v2.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":"universal-melgan-a-robust-neural-vocoder-for","repo_url":"https://github.com/huseinzol05/malaya-speech","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"universal-melgan-a-robust-neural-vocoder-for","repo_url":"https://github.com/avi33/universalmelgan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"text-to-speech","task_name":"Text to Speech"},{"task_slug":"text-to-speech-1","task_name":"text-to-speech"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dilated-convolution","method_name":"Dilated Convolution"},{"method_slug":"gan-hinge-loss","method_name":"GAN Hinge Loss"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"melgan","method_name":"MelGAN"},{"method_slug":"melgan-residual-block","method_name":"MelGAN Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"},{"method_slug":"weight-normalization","method_name":"Weight Normalization"},{"method_slug":"window-based-discriminator","method_name":"Window-based Discriminator"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2011.09631","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}