{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T10:12:56Z","timestamp":1780049576515,"version":"3.53.1"},"reference-count":47,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Samsung Electronics Company, Ltd.","award":["IO201211-08075-01"],"award-info":[{"award-number":["IO201211-08075-01"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE\/ACM Trans. Audio Speech Lang. Process."],"published-print":{"date-parts":[[2021]]},"DOI":"10.1109\/taslp.2021.3071662","type":"journal-article","created":{"date-parts":[[2021,4,7]],"date-time":"2021-04-07T20:29:07Z","timestamp":1617827347000},"page":"1626-1638","source":"Crossref","is-referenced-by-count":25,"title":["TutorNet: Towards Flexible Knowledge Distillation for End-to-End Speech Recognition"],"prefix":"10.1109","volume":"29","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8631-4489","authenticated-orcid":false,"given":"Ji Won","family":"Yoon","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6997-205X","authenticated-orcid":false,"given":"Hyeonseung","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hyung Yong","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Won Ik","family":"Cho","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0568-4902","authenticated-orcid":false,"given":"Nam Soo","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","first-page":"1","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"0","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref38","article-title":"Nemo: A toolkit for building ai applications using neural modules","author":"kuchaiev","year":"2019"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1296"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953077"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2017.2763455"},{"key":"ref30","article-title":"Aishell-2: Transforming mandarin ASR research into industrial scale","author":"du","year":"2018"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1456"},{"key":"ref36","article-title":"Mixed-precision training for NLP and speech recognition with OpenSeq2Seq","author":"kuchaiev","year":"2018"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref34","first-page":"1","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"0","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref10","first-page":"2654","article-title":"Do deep nets really need to be deep","author":"ba","year":"0","journal-title":"Proc Neural Inf Process Syst"},{"key":"ref40","article-title":"Stochastic gradient methods with layer-wise adaptive moments for training of deep networks","author":"ginsburg","year":"2019"},{"key":"ref11","first-page":"1","article-title":"FitNets: Hints for thin deep nets","author":"romero","year":"0","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref12","first-page":"1","article-title":"Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer","author":"zagoruyko","year":"0","journal-title":"Proc Int Conf Learn Representations"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.754"},{"key":"ref14","first-page":"4723","article-title":"Knowledge transfer with jacobian matching","author":"s","year":"0","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref15","first-page":"1910","article-title":"Learning small-size DNN with output-distribution-based criteria","author":"li","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1190"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953163"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953072"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-614"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178778"},{"key":"ref4","first-page":"173","article-title":"Deep speech 2: End-to-end speech recognition in english and mandarin","author":"amodei","year":"0","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1952"},{"key":"ref3","first-page":"1","article-title":"Sequence transduction with recurrent neural networks","author":"graves","year":"0","journal-title":"Proc ICML Workshop Representation Learn"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1819"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref5","article-title":"Wav2letter: An end-to-end convnet-based speech recognition system","author":"collobert","year":"2016"},{"key":"ref8","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"0","journal-title":"Proc Neural Inf Process Syst"},{"key":"ref7","first-page":"6124","article-title":"Quartznet: Deep automatic speech recognition with 1\ufffdd time-channel separable convolutions","author":"kriman","year":"0","journal-title":"Proc Int Conf Acoust Speech Signal Process"},{"key":"ref2","first-page":"577","article-title":"Attention-based models for speech recognition","author":"chorowski","year":"0","journal-title":"Proc Neural Inf Process Syst"},{"key":"ref9","first-page":"1","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"0","journal-title":"Proc Neural Inf Process Syst Workshop Deep Learn"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003936"},{"key":"ref20","first-page":"1","article-title":"Blending lstms into CNNs","author":"geras","year":"0","journal-title":"Proc Int Conf Learn Representations Workshop"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2019.2929859"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-546"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-911"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1162"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461995"},{"key":"ref41","first-page":"187","article-title":"Kenlm: Faster and smaller language model queries","author":"heafield","year":"0","journal-title":"Proc Workshop Statist Mach Transl"},{"key":"ref23","first-page":"604","article-title":"Acoustic modelling with CD-CTC-SMBR LSTM RNNS","author":"senior","year":"0","journal-title":"Proc IEEE Workshop Autom Speech Recognit Understanding"},{"key":"ref44","first-page":"1764","article-title":"Towards end-to-end speech recognition with recurrent neural networks","author":"graves","year":"0","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639629"},{"key":"ref43","article-title":"Adadelta: An adaptive learning rate method","author":"zeiler","year":"2012"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682671"}],"container-title":["IEEE\/ACM Transactions on Audio, Speech, and Language Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6570655\/9289074\/09398543.pdf?arnumber=9398543","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T14:53:59Z","timestamp":1652194439000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9398543\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":47,"URL":"https:\/\/doi.org\/10.1109\/taslp.2021.3071662","relation":{},"ISSN":["2329-9290","2329-9304"],"issn-type":[{"value":"2329-9290","type":"print"},{"value":"2329-9304","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021]]}}}