{"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/wav2letter-the-fastest-open-source-speech","title":"wav2letter++: The Fastest Open-source Speech Recognition System","arxiv_id":"1812.07625","date":"2018-12-18","proceeding":null,"authors":["Vineel Pratap","Awni Hannun","Qiantong Xu","Jeff Cai","Jacob Kahn","Gabriel Synnaeve","Vitaliy Liptchinsky","Ronan Collobert"],"abstract":"This paper introduces wav2letter++, the fastest open-source deep learning\nspeech recognition framework. wav2letter++ is written entirely in C++, and uses\nthe ArrayFire tensor library for maximum efficiency. Here we explain the\narchitecture and design of the wav2letter++ system and compare it to other\nmajor open-source speech recognition systems. In some cases wav2letter++ is\nmore than 2x faster than other optimized frameworks for training end-to-end\nneural networks for speech recognition. We also show that wav2letter++'s\ntraining times scale linearly to 64 GPUs, the highest we tested, for models\nwith 100 million parameters. High-performance frameworks enable fast iteration,\nwhich is often a crucial factor in successful research and model tuning on new\ndatasets and tasks.","url_abs":"http://arxiv.org/abs/1812.07625v1","url_pdf":"http://arxiv.org/pdf/1812.07625v1.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":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/bsridatta/wav2letter-Swedish","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/facebookresearch/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/flashlight/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/gcambara/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/jakeju/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/krantirk/Wav2letterPlus","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/lilei-John/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"wav2letter-the-fastest-open-source-speech","repo_url":"https://github.com/mailong25/wav2letter","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"speech-recognition","task_name":"Speech Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.07625","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}