{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T20:33:43Z","timestamp":1776890023705,"version":"3.51.2"},"reference-count":32,"publisher":"IEEE","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,12]]},"DOI":"10.1109\/slt.2018.8639629","type":"proceedings-article","created":{"date-parts":[[2019,2,14]],"date-time":"2019-02-14T18:36:34Z","timestamp":1550169394000},"page":"411-417","source":"Crossref","is-referenced-by-count":29,"title":["Improved Knowledge Distillation from Bi-Directional to Uni-Directional LSTM CTC for End-to-End Speech Recognition"],"prefix":"10.1109","author":[{"given":"Gakuto","family":"Kurata","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kartik","family":"Audhkhasi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref32","article-title":"Regularizing neural networks by penalizing confident output distributions","author":"pereyra","year":"2017","journal-title":"arXiv preprint arXiv 1701 00295"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2015.7404790"},{"key":"ref30","article-title":"Fast and accurate recurrent neural network acoustic models for speech recognition","author":"sak","year":"2015","journal-title":"arXiv preprint arXiv 1507 06947"},{"key":"ref10","doi-asserted-by":"crossref","first-page":"3707","DOI":"10.21437\/Interspeech.2017-1566","article-title":"Neural speech recognizer: Acoustic-to-word LSTM model for large vocabulary speech recognition","author":"soltau","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref11","first-page":"2188","article-title":"Gram-ctc: Automatic unit selection and target decomposition for sequence labelling","author":"liu","year":"2017","journal-title":"Proc ICML"},{"key":"ref12","article-title":"Building competitive direct acoustics-to-word models for English conversational speech recognition","author":"audhkhasi","year":"2017","journal-title":"arXiv preprint arXiv 1712 03133"},{"key":"ref13","first-page":"604","article-title":"Acoustic modelling with CD-CTC-sMBR LSTM RNNs","author":"sak","year":"2015","journal-title":"Proc ASRU"},{"key":"ref14","article-title":"Improved training for online end-to-end speech recognition systems","author":"kim","year":"2017","journal-title":"arXiv preprint arXiv 1711 03890"},{"key":"ref15","first-page":"63","article-title":"Reducing the computational complexity for whole word models","author":"soltau","year":"2017","journal-title":"Proc ASRU"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472152"},{"key":"ref17","first-page":"206","article-title":"Exploring neural transducers for end-to-end speech recognition","author":"battenberg","year":"2016","journal-title":"Proc ASRU"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461995"},{"key":"ref19","first-page":"2654","article-title":"Do deep nets really need to be deep?","author":"ba","year":"2014","journal-title":"Proc NIPS"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1139"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.2205597"},{"key":"ref27","first-page":"1764","article-title":"Towards end-to-end speech recognition with recurrent neural networks","author":"graves","year":"2014","journal-title":"Proc ICML"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/5.18626"},{"key":"ref6","doi-asserted-by":"crossref","DOI":"10.23915\/distill.00008","article-title":"Sequence modeling with CTC","author":"hannun","year":"2017","journal-title":"Distillation"},{"key":"ref29","doi-asserted-by":"crossref","first-page":"132","DOI":"10.21437\/Interspeech.2017-405","article-title":"English conversational telephone speech recognition by humans and machines","author":"saon","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref5","article-title":"Deep speech: Scaling up end-to-end speech recognition","author":"hannun","year":"2014","journal-title":"arXiv preprint arXiv 1412 5567"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref7","first-page":"577","article-title":"Attention-based models for speech recognition","author":"chorowski","year":"2015","journal-title":"Proc NIPS"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P17-1048"},{"key":"ref9","doi-asserted-by":"crossref","first-page":"959","DOI":"10.21437\/Interspeech.2017-546","article-title":"Direct acoustics-to-word models for English conversational speech recognition","author":"audhkhasi","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472618"},{"key":"ref20","article-title":"Distilling the knowledge in a neural network","author":"hinton","year":"2015","journal-title":"arXiv preprint arXiv 1503"},{"key":"ref22","doi-asserted-by":"crossref","first-page":"3697","DOI":"10.21437\/Interspeech.2017-614","article-title":"Efficient knowledge distillation from an ensemble of teachers","author":"fukuda","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref21","first-page":"1910","article-title":"Learning small-size DNN with output-distribution-based criteria","author":"li","year":"2014","journal-title":"Proc INTERSPEECH"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178778"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref26","article-title":"Generating sequences with recurrent neural networks","author":"graves","year":"2013","journal-title":"arXiv preprint arXiv 1308 0850"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"}],"event":{"name":"2018 IEEE Spoken Language Technology Workshop (SLT)","location":"Athens, Greece","start":{"date-parts":[[2018,12,18]]},"end":{"date-parts":[[2018,12,21]]}},"container-title":["2018 IEEE Spoken Language Technology Workshop (SLT)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8632666\/8639030\/08639629.pdf?arnumber=8639629","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,27]],"date-time":"2022-01-27T01:37:15Z","timestamp":1643247435000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/8639629\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,12]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/slt.2018.8639629","relation":{},"subject":[],"published":{"date-parts":[[2018,12]]}}}