{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T14:57:15Z","timestamp":1777129035101,"version":"3.51.4"},"reference-count":29,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,6,4]]},"DOI":"10.1109\/icassp49357.2023.10095616","type":"proceedings-article","created":{"date-parts":[[2023,5,5]],"date-time":"2023-05-05T17:28:30Z","timestamp":1683307710000},"page":"1-5","source":"Crossref","is-referenced-by-count":10,"title":["Joint Pre-Training with Speech and Bilingual Text for Direct Speech to Speech Translation"],"prefix":"10.1109","author":[{"given":"Kun","family":"Wei","sequence":"first","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xian,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Long","family":"Zhou","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ziqiang","family":"Zhang","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Liping","family":"Chen","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shujie","family":"Liu","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lei","family":"He","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinyu","family":"Li","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Furu","family":"Wei","sequence":"additional","affiliation":[{"name":"Microsoft Corporation"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref13","article-title":"Multilingual speech translation with efficient finetuning of pretrained models","author":"li","year":"2020"},{"key":"ref12","first-page":"12449","article-title":"wav2vec 2.0: A framework for self-supervised learning of speech representations","volume":"33","author":"baevski","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054626"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref11","article-title":"Translatotron 2: Robust direct speech-to-speech translation","author":"jia","year":"2021"},{"key":"ref10","article-title":"mslam: Massively multilingual joint pre-training for speech and text","author":"bapna","year":"2022"},{"key":"ref2","article-title":"Direct speech-to-speech translation with discrete units","author":"lee","year":"2021"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1951"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1997.599557"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TSA.2005.860774"},{"key":"ref19","article-title":"Listen and translate: A proof of concept for end-to-end speech-to-text translation","author":"b\u00e9rard","year":"2016"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2005-726"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00343"},{"key":"ref23","article-title":"Textless speech-to-speech translation on real data","author":"lee","year":"2021"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.emnlp-main.108"},{"key":"ref25","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"NIPS"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003853"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i16.17684"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/SLT48900.2021.9383496"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1145"},{"key":"ref27","first-page":"79","article-title":"Europarl: A parallel corpus for statistical machine translation","author":"koehn","year":"2005","journal-title":"Proceedings of Machine Translation Summit X Papers"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-6319"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10011"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.899"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10938"},{"key":"ref4","article-title":"Simple and effective unsupervised speech translation","author":"wang","year":"2022"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-11032"},{"key":"ref6","article-title":"Cvss corpus and massively multilingual speech-to-speech translation","author":"jia","year":"2022"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.80"}],"event":{"name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Rhodes Island, Greece","start":{"date-parts":[[2023,6,4]]},"end":{"date-parts":[[2023,6,10]]}},"container-title":["ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10094559\/10094560\/10095616.pdf?arnumber=10095616","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,20]],"date-time":"2023-11-20T19:00:19Z","timestamp":1700506819000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10095616\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,4]]},"references-count":29,"URL":"https:\/\/doi.org\/10.1109\/icassp49357.2023.10095616","relation":{},"subject":[],"published":{"date-parts":[[2023,6,4]]}}}