{"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/comsl-a-composite-speech-language-model-for-1","title":"ComSL: A Composite Speech-Language Model for End-to-End Speech-to-Text Translation","arxiv_id":"2305.14838","date":"2023-05-24","proceeding":"NeurIPS 2023 11","authors":["Chenyang Le","Yao Qian","Long Zhou","Shujie Liu","Yanmin Qian","Michael Zeng","Xuedong Huang"],"abstract":"Joint speech-language training is challenging due to the large demand for training data and GPU consumption, as well as the modality gap between speech and language. We present ComSL, a speech-language model built atop a composite architecture of public pretrained speech-only and language-only models and optimized data-efficiently for spoken language tasks. Particularly, we propose to incorporate cross-modality learning into transfer learning and conduct them simultaneously for downstream tasks in a multi-task learning manner. Our approach has demonstrated effectiveness in end-to-end speech-to-text translation tasks, achieving a new state-of-the-art average BLEU score of 31.5 on the multilingual speech to English text translation task for 21 languages, as measured on the public CoVoST2 evaluation set.","url_abs":"https://arxiv.org/abs/2305.14838v2","url_pdf":"https://arxiv.org/pdf/2305.14838v2.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":"comsl-a-composite-speech-language-model-for-1","repo_url":"https://github.com/nethermanpro/comsl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"speech-to-text","task_name":"Speech-to-Text"},{"task_slug":"speech-to-text-translation","task_name":"Speech-to-Text Translation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2305.14838","atlas_url":"https://app.syntology.ai/?focus=2305.14838","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}