{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,9,8]],"date-time":"2024-09-08T15:05:31Z","timestamp":1725807931254},"reference-count":35,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"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":[[2022,5,23]]},"DOI":"10.1109\/icassp43922.2022.9747879","type":"proceedings-article","created":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T19:50:34Z","timestamp":1651089034000},"page":"8112-8116","source":"Crossref","is-referenced-by-count":12,"title":["Improving The Latency And Quality Of Cascaded Encoders"],"prefix":"10.1109","author":[{"given":"Tara N.","family":"Sainath","sequence":"first","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Yanzhang","family":"He","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Arun","family":"Narayanan","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Rami","family":"Botros","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Weiran","family":"Wang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"David","family":"Qiu","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Chung-Cheng","family":"Chiu","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Rohit","family":"Prabhavalkar","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Alexander","family":"Gruenstein","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Anmol","family":"Gulati","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Bo","family":"Li","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"David","family":"Rybach","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Emmanuel","family":"Guzman","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Ian","family":"McGraw","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"James","family":"Qin","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Krzysztof","family":"Choromanski","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Qiao","family":"Liang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Robert","family":"David","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Ruoming","family":"Pang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Shuo-Yiin","family":"Chang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Trevor","family":"Strohman","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"W. Ronny","family":"Huang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Wei","family":"Han","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Yonghui","family":"Wu","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]},{"given":"Yu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}]}],"member":"263","reference":[{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413899"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-264"},{"key":"ref31","article-title":"Improving Tail Performance of a Deliberation E2E ASR Model Using a Large Text Corpus","author":"sainath","year":"2020","journal-title":"Proc INTERSPEECH"},{"key":"ref30","article-title":"Improving Wideband Speech Rcognition using Mixed-bandwidth Training Data in CD-DNN-HMM","author":"li","year":"2012","journal-title":"Proc SLT"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2012.6289079"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413803"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-212"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"article-title":"Fast transformer decoding: One write-head is all you need","year":"2019","author":"shazeer","key":"ref12"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413966"},{"article-title":"Fastformer: Additive attention can be all you need","year":"2021","author":"wu","key":"ref14"},{"article-title":"Efficient end-to-end speech recognition using performers in conformers","year":"2020","author":"wang","key":"ref15"},{"key":"ref16","article-title":"Rethinking attention with performers","author":"choromanski","year":"2021","journal-title":"9th International Conference on Learning Representations ICLR 2021 Virtual Event Austria May 3-7 2021"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9687874"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414560"},{"key":"ref19","article-title":"wav2vec 2.0: A framework for self-supervised learning of speech representations","author":"baevski","year":"2020","journal-title":"NeurIPS"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2013.6707758"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682336"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003913"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2846"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953075"},{"key":"ref29","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2017-1510","article-title":"Generation of Large-Scale Simulated Utterances in Virtual Rooms to Train Deep-Neural Networks for Far-Field Speech Recognition in Google Home","author":"kim","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462105"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1855"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003906"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-275"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-206"},{"key":"ref1","article-title":"Cascaded en-coders for unifying streaming and non-streaming ASR","author":"narayanan","year":"2021","journal-title":"Proc ICASSP"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1873"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461809"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688253"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054188"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054715"}],"event":{"name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2022,5,23]]},"location":"Singapore, Singapore","end":{"date-parts":[[2022,5,27]]}},"container-title":["ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9745891\/9746004\/09747879.pdf?arnumber=9747879","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,15]],"date-time":"2022-08-15T20:10:49Z","timestamp":1660594249000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9747879\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,23]]},"references-count":35,"URL":"https:\/\/doi.org\/10.1109\/icassp43922.2022.9747879","relation":{},"subject":[],"published":{"date-parts":[[2022,5,23]]}}}