{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,15]],"date-time":"2025-11-15T10:25:27Z","timestamp":1763202327652,"version":"3.28.0"},"reference-count":28,"publisher":"IEEE","license":[{"start":{"date-parts":[[2020,5,1]],"date-time":"2020-05-01T00:00:00Z","timestamp":1588291200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2020,5,1]],"date-time":"2020-05-01T00:00:00Z","timestamp":1588291200000},"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":[[2020,5]]},"DOI":"10.1109\/icassp40776.2020.9054116","type":"proceedings-article","created":{"date-parts":[[2020,4,9]],"date-time":"2020-04-09T16:21:13Z","timestamp":1586449273000},"page":"7059-7063","source":"Crossref","is-referenced-by-count":6,"title":["Independent Language Modeling Architecture for End-To-End ASR"],"prefix":"10.1109","author":[{"given":"Van Tung","family":"Pham","sequence":"first","affiliation":[{"name":"Nanyang Technological University,School of Computer Science and Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haihua","family":"Xu","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,School of Computer Science and Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yerbolat","family":"Khassanov","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,School of Computer Science and Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhiping","family":"Zeng","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,School of Computer Science and Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Eng Siong","family":"Chng","sequence":"additional","affiliation":[{"name":"Nanyang Technological University,School of Computer Science and Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chongjia","family":"Ni","sequence":"additional","affiliation":[{"name":"Alibaba Group,Machine Intelligence Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bin","family":"Ma","sequence":"additional","affiliation":[{"name":"Alibaba Group,Machine Intelligence Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haizhou","family":"Li","sequence":"additional","affiliation":[{"name":"National University of Singapore,Department of Electrical and Computer Engineering,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2017.2763455"},{"key":"ref11","article-title":"Catastrophic forgetting: still a problem for dnns","volume":"abs 1905 8077","author":"pf\u00fclb","year":"2019","journal-title":"CoRR"},{"key":"ref12","article-title":"Semi-supervised training for improving data efficiency in end-to-end speech synthesis","volume":"abs 1808 10128","author":"chung","year":"2018","journal-title":"CoRR"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.345"},{"key":"ref14","first-page":"389","article-title":"Endto-end speech recognition with word-based rnn language models","author":"hori","year":"2018","journal-title":"Proc of HLT"},{"journal-title":"On using monolingual corpora in neural machine translation","year":"2015","author":"gulcehre","key":"ref15"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683745"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-2456"},{"key":"ref18","first-page":"3079","article-title":"Semi-supervised sequence learning","author":"dai","year":"2015","journal-title":"Proc of NIPS"},{"key":"ref19","first-page":"5926","article-title":"MASS: masked sequence to sequence pre-training for language generation","author":"song","year":"2019","journal-title":"Proc of ICML"},{"key":"ref28","article-title":"Understanding the exploding gradient problem","volume":"abs 1211 5063","author":"pascanu","year":"2012","journal-title":"CoRR"},{"key":"ref4","first-page":"577","article-title":"Attention-based models for speech recognition","author":"chorowski","year":"2015","journal-title":"Proc of NIPS"},{"key":"ref27","article-title":"Adadelta: An adaptive learning rate method","volume":"abs 1212 5701","author":"zeiler","year":"2012","journal-title":"CoRR"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462105"},{"key":"ref6","doi-asserted-by":"crossref","first-page":"939","DOI":"10.21437\/Interspeech.2017-233","article-title":"A comparison of sequence-to-sequence models for speech recognition","author":"prabhavalkar","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref5","doi-asserted-by":"crossref","first-page":"523","DOI":"10.21437\/Interspeech.2017-343","article-title":"Towards better decoding and language model integration in sequence to sequence models","author":"chorowski","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref8","first-page":"173","article-title":"Deep speech 2 : End-to-end speech recognition in english and mandarin","year":"2015","journal-title":"Proc of ICML"},{"key":"ref7","first-page":"1764","article-title":"Towards end-to-end speech recognition with recurrent neural networks","author":"graves","year":"2014","journal-title":"Proc of ICML"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472618"},{"key":"ref9","first-page":"4835","article-title":"Joint ctc-attention based end-to-end speech recognition using multi-task learning","author":"kim","year":"2017","journal-title":"Proc of ICASSP"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1039"},{"key":"ref22","first-page":"724","article-title":"Hkust\/mts: A very large scale mandarin telephone speech corpus","author":"liu","year":"2006","journal-title":"proc of ISCSLP"},{"key":"ref21","article-title":"Sequence transduction with recurrent neural networks","volume":"abs 1211 3711","author":"graves","year":"2012","journal-title":"CoRR"},{"key":"ref24","first-page":"3586","article-title":"Audio augmentation for speech recognition","author":"ko","year":"2015","journal-title":"Proc of Interspeech"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1525"},{"key":"ref26","doi-asserted-by":"crossref","first-page":"949","DOI":"10.21437\/Interspeech.2017-1296","article-title":"Advances in joint ctc-attention based end-to-end speech recognition with a deep cnn encoder and rnnlm","author":"hori","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1456"}],"event":{"name":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2020,5,4]]},"location":"Barcelona, Spain","end":{"date-parts":[[2020,5,8]]}},"container-title":["ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9040208\/9052899\/09054116.pdf?arnumber=9054116","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,2]],"date-time":"2022-08-02T19:57:14Z","timestamp":1659470234000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9054116\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,5]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1109\/icassp40776.2020.9054116","relation":{},"subject":[],"published":{"date-parts":[[2020,5]]}}}