{"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/massively-multilingual-neural-grapheme-to","title":"Massively Multilingual Neural Grapheme-to-Phoneme Conversion","arxiv_id":"1708.01464","date":"2017-08-04","proceeding":"WS 2017 9","authors":["Ben Peters","Jon Dehdari","Josef van Genabith"],"abstract":"Grapheme-to-phoneme conversion (g2p) is necessary for text-to-speech and\nautomatic speech recognition systems. Most g2p systems are monolingual: they\nrequire language-specific data or handcrafting of rules. Such systems are\ndifficult to extend to low resource languages, for which data and handcrafted\nrules are not available. As an alternative, we present a neural\nsequence-to-sequence approach to g2p which is trained on\nspelling--pronunciation pairs in hundreds of languages. The system shares a\nsingle encoder and decoder across all languages, allowing it to utilize the\nintrinsic similarities between different writing systems. We show an 11%\nimprovement in phoneme error rate over an approach based on adapting\nhigh-resource monolingual g2p models to low-resource languages. Our model is\nalso much more compact relative to previous approaches.","url_abs":"http://arxiv.org/abs/1708.01464v1","url_pdf":"http://arxiv.org/pdf/1708.01464v1.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":"massively-multilingual-neural-grapheme-to","repo_url":"https://github.com/bpopeters/mg2p","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":null,"task_name":"Grapheme-to-Phoneme Conversion"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"text-to-speech","task_name":"Text to Speech"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"},{"task_slug":"text-to-speech-1","task_name":"text-to-speech"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.01464","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}