{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T15:24:53Z","timestamp":1775229893954,"version":"3.50.1"},"reference-count":34,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"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":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10890312","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T17:15:19Z","timestamp":1741799719000},"page":"1-5","source":"Crossref","is-referenced-by-count":3,"title":["EMMeTT: Efficient Multimodal Machine Translation Training"],"prefix":"10.1109","author":[{"given":"Piotr","family":"\u017belasko","sequence":"first","affiliation":[{"name":"NVIDIA"}]},{"given":"Zhehuai","family":"Chen","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Mengru","family":"Wang","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Daniel","family":"Galvez","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Oleksii","family":"Hrinchuk","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Shuoyang","family":"Ding","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Ke","family":"Hu","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Jagadeesh","family":"Balam","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Vitaly","family":"Lavrukhin","sequence":"additional","affiliation":[{"name":"NVIDIA"}]},{"given":"Boris","family":"Ginsburg","sequence":"additional","affiliation":[{"name":"NVIDIA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447553"},{"key":"ref2","article-title":"Salmonn: Towards generic hearing abilities for large language models","author":"Tang","year":"2023"},{"key":"ref3","article-title":"Qwen technical report","author":"Bai","year":"2023"},{"key":"ref4","article-title":"Audio flamingo: A novel audio language model with few-shot learning and dialogue abilities","author":"Kong","year":"2024"},{"key":"ref5","first-page":"23716","article-title":"Flamingo: a visual language model for few-shot learning","volume":"35","author":"Alayrac","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/SLT61566.2024.10832146"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.105"},{"key":"ref8","first-page":"12736","article-title":"Fused acoustic and text encoding for multimodal bilingual pretraining and speech translation","volume-title":"International Conference on Machine Learning","author":"Zheng"},{"key":"ref9","article-title":"Speech translation with large language models: An industrial practice","author":"Huang","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.iwslt-1.20"},{"key":"ref11","article-title":"Google USM: Scaling automatic speech recognition beyond 100 languages","author":"Zhang","year":"2023"},{"key":"ref12","article-title":"mslam: Massively multilingual joint pre-training for speech and text","author":"Bapna","year":"2022"},{"key":"ref13","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","author":"Raffel","year":"2020","journal-title":"Journal of Machine Learning Research"},{"key":"ref14","article-title":"Language models are few-shot learners","volume-title":"CoRR","author":"Brown","year":"2020"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00343"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.703"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref18","first-page":"110","article-title":"SOTASTREAM: A streaming approach to machine translation training","volume-title":"Proceedings of the 3rd Workshop for Natural Language Processing Open Source Software (NLP-OSS 2023)","author":"Post"},{"key":"ref19","article-title":"Lhotse: a speech data representation library for the modern deep learning ecosystem","volume-title":"Proceedings of NeurIPS Workshop on Data-Centric AI","author":"\u017belasko"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-2294"},{"key":"ref21","article-title":"NVIDIA Megatron NMT en-to-any 500M"},{"key":"ref22","article-title":"NVIDIA Megatron NMT any-to-en 500M"},{"key":"ref23","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2023.findings-acl.597","article-title":"Leveraging synthetic targets for machine translation","author":"Mittal","year":"2023"},{"key":"ref24","article-title":"Culturax: A cleaned, enormous, and multilingual dataset for large language models in 167 languages","author":"Nguyen","year":"2023"},{"key":"ref25","first-page":"291","article-title":"Bifixer and bicleaner: two open-source tools to clean your parallel data","volume-title":"Proceedings of the 22nd Annual Conference of the European Association for Machine Translation","author":"Ram\u00edrez-S\u00e1nchez"},{"key":"ref26","article-title":"Nemo: a toolkit for building ai applications using neural modules","author":"Kuchaiev","year":"2019"},{"key":"ref27","article-title":"Tinyllama: An open-source small language model","author":"Zhang","year":"2024"},{"key":"ref28","article-title":"Adam: A method for stochastic optimization","volume-title":"Proceedings of ICLR","author":"Kingma"},{"key":"ref29","article-title":"NVIDIA Riva Megatron NMT any-to-any 1B"},{"key":"ref30","article-title":"Root mean square layer normalization","volume-title":"Proceedings of NeurIPS","author":"Zhang"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/SLT54892.2023.10023141"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00474"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-6319"},{"key":"ref34","article-title":"Seamless: Multilingual expressive and streaming speech translation","author":"Barrault","year":"2023"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Hyderabad, India","start":{"date-parts":[[2025,4,6]]},"end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10890312.pdf?arnumber=10890312","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T05:24:50Z","timestamp":1774416290000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10890312\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":34,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10890312","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}