{"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/language-and-noise-transfer-in-speech","title":"Language and Noise Transfer in Speech Enhancement Generative Adversarial Network","arxiv_id":"1712.06340","date":"2017-12-18","proceeding":null,"authors":["Santiago Pascual","Maruchan Park","Joan Serrà","Antonio Bonafonte","Kang-Hun Ahn"],"abstract":"Speech enhancement deep learning systems usually require large amounts of\ntraining data to operate in broad conditions or real applications. This makes\nthe adaptability of those systems into new, low resource environments an\nimportant topic. In this work, we present the results of adapting a speech\nenhancement generative adversarial network by finetuning the generator with\nsmall amounts of data. We investigate the minimum requirements to obtain a\nstable behavior in terms of several objective metrics in two very different\nlanguages: Catalan and Korean. We also study the variability of test\nperformance to unseen noise as a function of the amount of different types of\nnoise available for training. Results show that adapting a pre-trained English\nmodel with 10 min of data already achieves a comparable performance to having\ntwo orders of magnitude more data. They also demonstrate the relative stability\nin test performance with respect to the number of training noise types.","url_abs":"http://arxiv.org/abs/1712.06340v1","url_pdf":"http://arxiv.org/pdf/1712.06340v1.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":"language-and-noise-transfer-in-speech","repo_url":"https://github.com/develooper1994/MasterThesis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"language-and-noise-transfer-in-speech","repo_url":"https://github.com/rickyHong/segan-pytorch-repl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"language-and-noise-transfer-in-speech","repo_url":"https://github.com/santi-pdp/segan_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"speech-enhancement","task_name":"Speech Enhancement"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}