{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T19:04:20Z","timestamp":1776884660764,"version":"3.51.2"},"reference-count":32,"publisher":"IEEE","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,1,19]]},"DOI":"10.1109\/slt48900.2021.9383521","type":"proceedings-article","created":{"date-parts":[[2021,3,25]],"date-time":"2021-03-25T20:46:54Z","timestamp":1616705214000},"page":"89-96","source":"Crossref","is-referenced-by-count":30,"title":["On The Usefulness of Self-Attention for Automatic Speech Recognition with Transformers"],"prefix":"10.1109","author":[{"given":"Shucong","family":"Zhang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Erfan","family":"Loweimi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Peter","family":"Bell","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Steve","family":"Renals","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref32","first-page":"630","article-title":"Identity mappings in deep residual networks","author":"he","year":"2016","journal-title":"European Conference on Computer Vision"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref30","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2015","journal-title":"ICLR 2015 International Conference on Learning Representations 2015"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/72.279181"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref12","article-title":"Empirical evaluation of gated recurrent neural networks on sequence modeling","author":"chung","year":"2014"},{"key":"ref13","article-title":"Neural machine translation by jointly learning to align and translate","author":"bahdanau","year":"2015","journal-title":"ICLR 2015 International Conference on Learning Representations 2015"},{"key":"ref14","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref15","first-page":"431","article-title":"Local monotonic attention mechanism for end-to-end speech and language processing","author":"tjandra","year":"2017","journal-title":"Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1 Long Papers)"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953075"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682224"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.3115\/1075527.1075614"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1456"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462497"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1162"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2702"},{"key":"ref6","article-title":"Transformer-based acoustic modeling for hybrid speech recognition","author":"wang","year":"2019"},{"key":"ref29","article-title":"Automatic differentiation in pytorch","author":"paszke","year":"2017","journal-title":"NIPS 2017 Workshop on Autodiff"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9004025"},{"key":"ref8","article-title":"Improving transformer-based end-to-end speech recognition with connectionist temporal classification and language model integration","author":"nakatani","year":"2019","journal-title":"Interspeech 2019"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003750"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1107"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2638"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462506"},{"key":"ref20","first-page":"14014","article-title":"Are sixteen heads really better than one","author":"michel","year":"2019","journal-title":"NeurIPS 2019 Thirty-third Conference on Neural Information Processing Systems"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178838"},{"key":"ref21","article-title":"How much self-attention do we need? trading attention for feed-forward layers","author":"irie","year":"2020","journal-title":"2020 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)"},{"key":"ref24","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfit-ting","volume":"15","author":"srivastava","year":"2014","journal-title":"Journal of Machine Learning Research"},{"key":"ref23","article-title":"Layer normalization","author":"ba","year":"2016"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854049"},{"key":"ref25","article-title":"The Kaldi speech recognition toolkit","author":"povey","year":"2011","journal-title":"IEEE ASRU"}],"event":{"name":"2021 IEEE Spoken Language Technology Workshop (SLT)","location":"Shenzhen, China","start":{"date-parts":[[2021,1,19]]},"end":{"date-parts":[[2021,1,22]]}},"container-title":["2021 IEEE Spoken Language Technology Workshop (SLT)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9383468\/9383452\/09383521.pdf?arnumber=9383521","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,5,31]],"date-time":"2021-05-31T21:30:39Z","timestamp":1622496639000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9383521\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,1,19]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/slt48900.2021.9383521","relation":{},"subject":[],"published":{"date-parts":[[2021,1,19]]}}}