{"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/lightweight-adapter-tuning-for-multilingual","title":"Lightweight Adapter Tuning for Multilingual Speech Translation","arxiv_id":"2106.01463","date":"2021-06-02","proceeding":"ACL 2021 5","authors":["Hang Le","Juan Pino","Changhan Wang","Jiatao Gu","Didier Schwab","Laurent Besacier"],"abstract":"Adapter modules were recently introduced as an efficient alternative to fine-tuning in NLP. Adapter tuning consists in freezing pretrained parameters of a model and injecting lightweight modules between layers, resulting in the addition of only a small number of task-specific trainable parameters. While adapter tuning was investigated for multilingual neural machine translation, this paper proposes a comprehensive analysis of adapters for multilingual speech translation (ST). Starting from different pre-trained models (a multilingual ST trained on parallel data or a multilingual BART (mBART) trained on non-parallel multilingual data), we show that adapters can be used to: (a) efficiently specialize ST to specific language pairs with a low extra cost in terms of parameters, and (b) transfer from an automatic speech recognition (ASR) task and an mBART pre-trained model to a multilingual ST task. Experiments show that adapter tuning offer competitive results to full fine-tuning, while being much more parameter-efficient.","url_abs":"https://arxiv.org/abs/2106.01463v2","url_pdf":"https://arxiv.org/pdf/2106.01463v2.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":"lightweight-adapter-tuning-for-multilingual","repo_url":"https://github.com/formiel/fairseq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"lightweight-adapter-tuning-for-multilingual","repo_url":"https://github.com/formiel/fairseq/blob/master/examples/speech_to_text/docs/adapters.md","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","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":"machine-translation","task_name":"Machine Translation"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-to-text-translation","task_name":"Speech-to-Text Translation"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"adapter","method_name":"Adapter"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bart","method_name":"BART"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"mbart","method_name":"mBART"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-to-text-translation-on-must-c-1","task":"Speech-to-Text Translation","dataset":"MuST-C","model":"Transformer with Adapters","rank_in_archive_order":1,"of":2,"metrics":{"SacreBLEU":"26.61"},"uses_additional_data":false},{"leaderboard":"/sota/speech-to-text-translation-on-must-c-en-de","task":"Speech-to-Text Translation","dataset":"MuST-C EN->DE","model":"Transformer with Adapters","rank_in_archive_order":4,"of":8,"metrics":{"Case-sensitive sacreBLEU":"24.63"},"uses_additional_data":false},{"leaderboard":"/sota/speech-to-text-translation-on-must-c-en-es","task":"Speech-to-Text Translation","dataset":"MuST-C EN->ES","model":"Transformer with Adapters","rank_in_archive_order":1,"of":5,"metrics":{"Case-sensitive sacreBLEU":"28.73"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.01463","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}