{"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/speech-waveform-synthesis-from-mfcc-sequences","title":"Speech waveform synthesis from MFCC sequences with generative adversarial networks","arxiv_id":"1804.00920","date":"2018-04-03","proceeding":null,"authors":["Lauri Juvela","Bajibabu Bollepalli","Xin Wang","Hirokazu Kameoka","Manu Airaksinen","Junichi Yamagishi","Paavo Alku"],"abstract":"This paper proposes a method for generating speech from filterbank mel\nfrequency cepstral coefficients (MFCC), which are widely used in speech\napplications, such as ASR, but are generally considered unusable for speech\nsynthesis. First, we predict fundamental frequency and voicing information from\nMFCCs with an autoregressive recurrent neural net. Second, the spectral\nenvelope information contained in MFCCs is converted to all-pole filters, and a\npitch-synchronous excitation model matched to these filters is trained.\nFinally, we introduce a generative adversarial network -based noise model to\nadd a realistic high-frequency stochastic component to the modeled excitation\nsignal. The results show that high quality speech reconstruction can be\nobtained, given only MFCC information at test time.","url_abs":"http://arxiv.org/abs/1804.00920v1","url_pdf":"http://arxiv.org/pdf/1804.00920v1.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":"speech-waveform-synthesis-from-mfcc-sequences","repo_url":"https://github.com/ljuvela/ResGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"speech-synthesis","task_name":"Speech Synthesis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.00920","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}