{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,10,29]],"date-time":"2024-10-29T10:55:41Z","timestamp":1730199341764,"version":"3.28.0"},"reference-count":48,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,12,13]],"date-time":"2021-12-13T00:00:00Z","timestamp":1639353600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,12,13]],"date-time":"2021-12-13T00:00:00Z","timestamp":1639353600000},"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":[[2021,12,13]]},"DOI":"10.1109\/asru51503.2021.9688028","type":"proceedings-article","created":{"date-parts":[[2022,2,3]],"date-time":"2022-02-03T20:31:00Z","timestamp":1643920260000},"page":"518-525","source":"Crossref","is-referenced-by-count":10,"title":["Kaizen: Continuously Improving Teacher Using Exponential Moving Average for Semi-Supervised Speech Recognition"],"prefix":"10.1109","author":[{"given":"Vimal","family":"Manohar","sequence":"first","affiliation":[{"name":"Facebook AI"}]},{"given":"Tatiana","family":"Likhomanenko","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Qiantong","family":"Xu","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Wei-Ning","family":"Hsu","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Ronan","family":"Collobert","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Yatharth","family":"Saraf","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Geoffrey","family":"Zweig","sequence":"additional","affiliation":[{"name":"Facebook AI"}]},{"given":"Abdelrahman","family":"Mohamed","sequence":"additional","affiliation":[{"name":"Facebook AI"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-2012"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1423"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-595"},{"journal-title":"Low latency modeling of temporal contexts for speech recognition","year":"2017","author":"peddinti","key":"ref32"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-129"},{"key":"ref30","article-title":"End-to-end asr: from supervised to semi-supervised learning with modern architectures","author":"synnaeve","year":"0","journal-title":"ICML SAS Workshop"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-647"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(90)90044-L"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688056"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TSA.2004.838537"},{"key":"ref40","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2014","journal-title":"ArXiv Preprint"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1470"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1800"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1280"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-740"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052942"},{"key":"ref17","article-title":"Temporal ensembling for semi-supervised learning","author":"laine","year":"2016","journal-title":"ArXiv Preprint"},{"key":"ref18","article-title":"Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results","volume":"30","author":"tarvainen","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref28","article-title":"Wav2letter: an end-to-end convnet-based speech recognition system","author":"collobert","year":"2016","journal-title":"ArXiv Preprint"},{"key":"ref4","article-title":"Model compression","author":"buciluundefined","year":"0","journal-title":"Proceedings of the 12th ACM SIGKDD"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-711"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054295"},{"key":"ref6","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"0","journal-title":"Deep Learning and Representation Learning Workshop NIPS"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3107"},{"key":"ref5","article-title":"Do deep nets really need to be deep?","author":"ba","year":"2014","journal-title":"NIPS"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462331"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-911"},{"key":"ref2","first-page":"6704","article-title":"Deep neural network features and semi-supervised training for low resource speech recognition","author":"samuel","year":"0","journal-title":"Proc ICASSP"},{"key":"ref9","article-title":"Semi-supervised speech recognition via local prior matching","author":"hsu","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref1","article-title":"Utilizing untranscribed training data to improve performance","author":"zavaliagkos","year":"0","journal-title":"DARPA Broadcast News Transcription and Understanding Workshop"},{"key":"ref46","article-title":"Mixed precision training","author":"narang","year":"0","journal-title":"Int Conf on Learning Representations"},{"key":"ref20","first-page":"21271","article-title":"Bootstrap your own latent -a new approach to self-supervised learning","volume":"33","author":"grill","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref45","first-page":"2121","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"duchi","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680"},{"key":"ref47","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":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-571"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054345"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003972"},{"key":"ref41","first-page":"5998","article-title":"Attention is all you need","author":"ashish","year":"0","journal-title":"Advances in neural information processing systems"},{"key":"ref23","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"srivastava","year":"2014","journal-title":"The Journal of Machine Learning Research"},{"key":"ref44","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"0","journal-title":"ICLRE"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1917"},{"key":"ref43","article-title":"A neural network for speaker-independent isolated word recognition","author":"yamaguchi","year":"0","journal-title":"Intl Conference on Spoken Language Processing"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177729694"}],"event":{"name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","start":{"date-parts":[[2021,12,13]]},"location":"Cartagena, Colombia","end":{"date-parts":[[2021,12,17]]}},"container-title":["2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9687821\/9687855\/09688028.pdf?arnumber=9688028","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,16]],"date-time":"2022-05-16T20:42:09Z","timestamp":1652733729000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9688028\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,13]]},"references-count":48,"URL":"https:\/\/doi.org\/10.1109\/asru51503.2021.9688028","relation":{},"subject":[],"published":{"date-parts":[[2021,12,13]]}}}