{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T18:59:02Z","timestamp":1776884342135,"version":"3.51.2"},"reference-count":29,"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"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,5,23]]},"DOI":"10.1109\/icassp43922.2022.9747216","type":"proceedings-article","created":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T19:50:34Z","timestamp":1651089034000},"page":"9256-9260","source":"Crossref","is-referenced-by-count":15,"title":["The DKU Audio-Visual Wake Word Spotting System for the 2021 MISP Challenge"],"prefix":"10.1109","author":[{"given":"Ming","family":"Cheng","sequence":"first","affiliation":[{"name":"School of Computer Science, Wuhan University,Wuhan,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haoxu","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University,Wuhan,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yechen","family":"Wang","sequence":"additional","affiliation":[{"name":"Data Science Research Center, Duke Kunshan University,Kunshan,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ming","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University,Wuhan,China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414777"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1412"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054039"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2018.8451096"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746683"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.510"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00675"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.617"},{"key":"ref19","first-page":"5533","article-title":"Learning spatiotemporal representation with pseudo-3d residual networks","author":"qiu","year":"2017","journal-title":"Proc ICCV 2017"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00482"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/PROC.1973.9030"},{"key":"ref27","first-page":"2825","article-title":"Scikitlearn: Machine learning in python","volume":"12","author":"pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1991.150338"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1363"},{"key":"ref29","first-page":"1","article-title":"Lrw-1000: A naturally-distributed large-scale benchmark for lip reading in the wild","author":"yang","year":"2019","journal-title":"Proc FG 2019"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-352"},{"key":"ref8","first-page":"220","article-title":"An application of recurrent neural networks to discriminative keyword spotting","author":"fern\u00e1ndez","year":"2007","journal-title":"Proc ICANN 2007"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1058"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/29.103088"},{"key":"ref9","doi-asserted-by":"crossref","first-page":"252","DOI":"10.1016\/j.specom.2012.08.006","article-title":"Keyword spotting exploiting long short-term memory","volume":"55","author":"woellmer","year":"2013","journal-title":"Speech Communication"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1990.115555"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1437"},{"key":"ref22","author":"oppenheim","year":"1997","journal-title":"Signals & Systems"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2017.373"},{"key":"ref24","article-title":"When does label smoothing helpƒ","author":"m\u00fcller","year":"2019","journal-title":"arXiv preprint arXiv 1906 02916"},{"key":"ref23","article-title":"Torchaudio: Building blocks for audio and speech processing","author":"yang","year":"2021","journal-title":"arXiv preprint arXiv 2110 13189"},{"key":"ref26","article-title":"Retinaface: Single-stage dense face localisation in the wild","author":"deng","year":"2019","journal-title":"ArXiv"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2009.2025790"}],"event":{"name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Singapore, Singapore","start":{"date-parts":[[2022,5,23]]},"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\/09747216.pdf?arnumber=9747216","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,15]],"date-time":"2022-08-15T20:10:05Z","timestamp":1660594205000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9747216\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,23]]},"references-count":29,"URL":"https:\/\/doi.org\/10.1109\/icassp43922.2022.9747216","relation":{},"subject":[],"published":{"date-parts":[[2022,5,23]]}}}