{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T18:15:17Z","timestamp":1777486517660,"version":"3.51.4"},"reference-count":103,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"6","license":[{"start":{"date-parts":[[2022,10,1]],"date-time":"2022-10-01T00:00:00Z","timestamp":1664582400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2022,10,1]],"date-time":"2022-10-01T00:00:00Z","timestamp":1664582400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,10,1]],"date-time":"2022-10-01T00:00:00Z","timestamp":1664582400000},"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":["IEEE J. Sel. Top. Signal Process."],"published-print":{"date-parts":[[2022,10]]},"DOI":"10.1109\/jstsp.2022.3182537","type":"journal-article","created":{"date-parts":[[2022,6,13]],"date-time":"2022-06-13T20:54:09Z","timestamp":1655153649000},"page":"1519-1532","source":"Crossref","is-referenced-by-count":124,"title":["BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition"],"prefix":"10.1109","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9505-1833","authenticated-orcid":false,"given":"Yu","family":"Zhang","sequence":"first","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1919-0460","authenticated-orcid":false,"given":"Daniel S.","family":"Park","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4201-9645","authenticated-orcid":false,"given":"Wei","family":"Han","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"James","family":"Qin","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Anmol","family":"Gulati","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6729-5988","authenticated-orcid":false,"given":"Joel","family":"Shor","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Aren","family":"Jansen","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Yuanzhong","family":"Xu","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Yanping","family":"Huang","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Shibo","family":"Wang","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Zongwei","family":"Zhou","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6711-3603","authenticated-orcid":false,"given":"Bo","family":"Li","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3132-4898","authenticated-orcid":false,"given":"Min","family":"Ma","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"William","family":"Chan","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Jiahui","family":"Yu","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Yongqiang","family":"Wang","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Liangliang","family":"Cao","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0866-2223","authenticated-orcid":false,"given":"Khe Chai","family":"Sim","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8049-2345","authenticated-orcid":false,"given":"Bhuvana","family":"Ramabhadran","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4126-6556","authenticated-orcid":false,"given":"Tara N.","family":"Sainath","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Francoise","family":"Beaufays","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8631-2424","authenticated-orcid":false,"given":"Zhifeng","family":"Chen","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Quoc V.","family":"Le","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Chung-Cheng","family":"Chiu","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]},{"given":"Ruoming","family":"Pang","sequence":"additional","affiliation":[{"name":"Google, USA"}]},{"given":"Yonghui","family":"Wu","sequence":"additional","affiliation":[{"name":"Google, Inc., USA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462037"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-2341"},{"key":"ref3","article-title":"Representation learning with contrastive predictive coding","author":"Oord","year":"2018"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-1473"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2019.2938863"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-1873"},{"key":"ref7","article-title":"vq-wav2vec: Self-supervised learning of discrete speech representations","author":"Baevski","year":"2019"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053176"},{"key":"ref9","article-title":"Effectiveness of self-supervised pre-training for speech recognition","author":"Baevski","year":"2019"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054548"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.106"},{"key":"ref12","article-title":"Wav2vec 2.0: A framework for self-supervised learning of speech representations","author":"Baevski","year":"2020"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688018"},{"key":"ref14","first-page":"301","article-title":"Utilizing untranscribed training data to improve performance","volume-title":"Proc. DARPA Broadcast News Transcription Understanding Workshop","author":"Zavaliagkos","year":"1998"},{"key":"ref15","first-page":"150","article-title":"Lightly supervised acoustic model training","volume-title":"Proc. ISCA ITRW","author":"Lamel","year":"2000"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2009-86"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638959"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682172"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054295"},{"key":"ref20","article-title":"End-to-end ASR: From supervised to semi-supervised learning with modern architectures","author":"Synnaeve","year":"2019"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683690"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/SLT48900.2021.9383552"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1470"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1800"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-677"},{"key":"ref26","article-title":"Pushing the limits of semi-supervised learning for automatic speech recognition","author":"Zhang","year":"2020"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052942"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414641"},{"key":"ref30","article-title":"Speechstew: Simply mix all available speech recognition data to train one large neural network","author":"Chan","year":"2021"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413899"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2020-3015"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-24797-2"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9687871"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1242"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952261"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.1965.1053799"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.3115\/981658.981684"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.3115\/1119355.1119369"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01070"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-2680"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053205"},{"key":"ref44","article-title":"Twin networks: Matching the future for sequence generation","author":"Serdyuk","year":"2017"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2018-2517"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1407"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461995"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639629"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2442"},{"key":"ref50","article-title":"Universal ASR: Unify and improve streaming ASR with full-context modeling","author":"Yu","year":"2020"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413692"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-637"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639610"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1758"},{"key":"ref56","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"issue":"8","key":"ref57","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref58","article-title":"Mesh-tensorflow: Deep learning for supercomputers","author":"Shazeer","year":"2018"},{"key":"ref59","article-title":"Megatron-LM: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00024"},{"key":"ref61","article-title":"Automatic cross-replica sharding of weight update in data-parallel training","author":"Xu","year":"2020"},{"key":"ref62","article-title":"GShard: Scaling giant models with conditional computation and automatic sharding","volume":"abs\/2006.16668","author":"Lepikhin","year":"2020"},{"key":"ref63","article-title":"GSPMD: General and scalable parallelization for ML computation graphs","volume":"abs\/2105.04663","author":"Xu","year":"2021"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/p19-1285"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2012.6289079"},{"key":"ref66","first-page":"5998","article-title":"Attention is all you need","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vaswani","year":"2017"},{"key":"ref67","first-page":"4603","article-title":"Adafactor: Adaptive learning rates with sublinear memory cost","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","volume":"80","author":"Shazeer","year":"2018"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/SLT48900.2021.9383518"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2013.6707758"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.21437\/CHiME.2020-1"},{"key":"ref71","article-title":"On using monolingual corpora in neural machine translation","author":"Gulcehre","year":"2015"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-102"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-211"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1007\/11677482_3"},{"key":"ref75","article-title":"Common voice: A massively-multilingual speech corpus","author":"Ardila","year":"2019"},{"key":"ref76","first-page":"125","article-title":"TED-LIUM: An automatic speech recognition dedicated corpus","volume-title":"LREC","author":"Rousseau","year":"2012"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-99579-3_21"},{"key":"ref78","first-page":"1","article-title":"The kaldi speech recognition toolkit","volume-title":"Proc. IEEE Workshop Autom. Speech Recognit. Understanding","author":"Povey","year":"2011"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.21437\/CHiME.2018-8"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.21437\/chime.2020\u20139"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682336"},{"key":"ref82","article-title":"Feature learning with raw-waveform CLDNNS for voice activity detection","author":"Candil","year":"2016"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-350"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053896"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1109\/TAFFC.2014.2336244"},{"key":"ref86","first-page":"53","article-title":"Speaker-dependent audio-visual emotion recognition","author":"Haq","year":"2009","journal-title":"AVSP"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-950"},{"key":"ref88","article-title":"Voxforge","author":"MacLean","year":"2018"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-2070"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-0032"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2806390"},{"key":"ref92","article-title":"Dementiabank database guide","author":"Boller","year":"2005"},{"key":"ref93","article-title":"Speech commands: A dataset for limited-vocabulary speech recognition","author":"Warden","year":"2018"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-34255-5_17"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3078715"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1109\/ACII.2019.8925444"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2857"},{"key":"ref98","article-title":"On the Importance of Normative Data in Speech-Based Assessment","volume":"abs\/1712.00069","author":"Noorian","year":"2017"},{"key":"ref99","article-title":"Yamnet","author":"Plakal","year":"2020"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1913"},{"key":"ref101","first-page":"2825","article-title":"Scikit-learn: Machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461684"},{"key":"ref103","article-title":"Multi-format contrastive learning of audio representations","author":"Wang","year":"2021"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00129"}],"container-title":["IEEE Journal of Selected Topics in Signal Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/4200690\/9923627\/09795080.pdf?arnumber=9795080","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T03:56:20Z","timestamp":1706759780000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9795080\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10]]},"references-count":103,"journal-issue":{"issue":"6"},"URL":"https:\/\/doi.org\/10.1109\/jstsp.2022.3182537","relation":{},"ISSN":["1932-4553","1941-0484"],"issn-type":[{"value":"1932-4553","type":"print"},{"value":"1941-0484","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,10]]}}}