{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T20:23:52Z","timestamp":1776889432683,"version":"3.51.2"},"reference-count":25,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,12,6]],"date-time":"2025-12-06T00:00:00Z","timestamp":1764979200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,6]],"date-time":"2025-12-06T00:00:00Z","timestamp":1764979200000},"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":[[2025,12,6]]},"DOI":"10.1109\/asru65441.2025.11434640","type":"proceedings-article","created":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T19:48:04Z","timestamp":1775159284000},"page":"1-7","source":"Crossref","is-referenced-by-count":2,"title":["Training and Inference Efficiency of Encoder-Decoder Speech Models"],"prefix":"10.1109","author":[{"given":"Piotr","family":"\u017belasko","sequence":"first","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Kunal","family":"Dhawan","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Daniel","family":"Galvez","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Krishna C.","family":"Puvvada","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Ankita","family":"Pasad","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Travis","family":"Bartley","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Nithin Rao","family":"Koluguri","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Ke","family":"Hu","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Vitaly","family":"Lavrukhin","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Jagadeesh","family":"Balam","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]},{"given":"Boris","family":"Ginsburg","sequence":"additional","affiliation":[{"name":"NVIDIA,USA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref2","article-title":"Robust speech recognition via large-scale weak supervision","author":"Radford"},{"key":"ref3","article-title":"Seamless: Multilingual expressive and streaming speech translation","author":"Barrault","year":"2023","journal-title":"arXiv preprint arXiv:2312.05187"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389676"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2024-1194"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-2294"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/DSMP.2016.7583516"},{"key":"ref10","article-title":"A comprehensive study of batch construction strategies for recurrent neural networks in mxnet","author":"Doetsch"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10890312"},{"key":"ref12","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":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref15","first-page":"2685","article-title":"COMET: A neural framework for MT evaluation","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing","author":"Rei"},{"key":"ref16","article-title":"Fleurs: Few-shot learning evaluation of universal representations of speech","author":"Conneau","year":"2022","journal-title":"arXiv preprint arXiv:2205.12446"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-2027"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/w18-6319"},{"key":"ref19","article-title":"Common voice: A massively-multilingual speech corpus","author":"Ardila"},{"key":"ref20","article-title":"Deep encoder, shallow decoder: Reevaluating non-autoregressive machine translation","volume-title":"9th International Conference on Learning Representations, ICLR 2021","author":"Kasai"},{"key":"ref21","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"key":"ref22","article-title":"Neural machine translation by jointly learning to align and translate","author":"Bahdanau","year":"2015"},{"key":"ref23","article-title":"Distil-whisper: Robust knowledge distillation via large-scale pseudo labelling","author":"Gandhi"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389701"},{"key":"ref25","article-title":"Sequence transduction with recurrent neural networks","volume-title":"Internation Conference on Machine Learning (ICML) Workshop on Representation Learning","author":"Graves"},{"key":"ref26","article-title":"An empirical model of large-batch training","author":"McCandlish"},{"issue":"112","key":"ref27","first-page":"1","article-title":"Measuring the effects of data parallelism on neural network training","volume":"20","author":"Shallue","year":"2019","journal-title":"Journal of Machine Learning Research"},{"key":"ref28","article-title":"How does critical batch size scale in pre-training?","author":"Zhang"}],"event":{"name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,12,6]]},"end":{"date-parts":[[2025,12,10]]}},"container-title":["2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11434577\/11433836\/11434640.pdf?arnumber=11434640","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T04:57:42Z","timestamp":1775192262000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11434640\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,6]]},"references-count":25,"URL":"https:\/\/doi.org\/10.1109\/asru65441.2025.11434640","relation":{},"subject":[],"published":{"date-parts":[[2025,12,6]]}}}