{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,11]],"date-time":"2026-02-11T17:17:49Z","timestamp":1770830269484,"version":"3.50.1"},"reference-count":26,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"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":[[2023,6,4]]},"DOI":"10.1109\/icassp49357.2023.10096858","type":"proceedings-article","created":{"date-parts":[[2023,5,5]],"date-time":"2023-05-05T13:28:30Z","timestamp":1683293310000},"page":"1-5","source":"Crossref","is-referenced-by-count":9,"title":["VE-KWS: Visual Modality Enhanced End-to-End Keyword Spotting"],"prefix":"10.1109","author":[{"given":"Ao","family":"Zhang","sequence":"first","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xi’an,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"He","family":"Wang","sequence":"additional","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xi’an,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pengcheng","family":"Guo","sequence":"additional","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xi’an,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yihui","family":"Fu","sequence":"additional","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xi’an,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lei","family":"Xie","sequence":"additional","affiliation":[{"name":"Northwestern Polytechnical University,Audio, Speech and Language Processing Group (ASLP@NPU), School of Computer Science,Xi’an,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yingying","family":"Gao","sequence":"additional","affiliation":[{"name":"China Mobile Research Institute,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shilei","family":"Zhang","sequence":"additional","affiliation":[{"name":"China Mobile Research Institute,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junlan","family":"Feng","sequence":"additional","affiliation":[{"name":"China Mobile Research Institute,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CISP-BMEI51763.2020.9263564"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854370"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414777"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053009"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682682"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1230"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1521"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747025"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW56347.2022.00278"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-741"},{"key":"ref11","article-title":"Seeing wake words: Audio-visual keyword spotting","volume-title":"Proc. BMVC","author":"Momeni"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746683"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747216"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746762"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2018.2881912"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2266"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3181"},{"key":"ref18","first-page":"276","article-title":"Advances in online audiovisual meeting transcription","volume-title":"Proc. ASRU","author":"Yoshioka"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.590"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2017.373"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1437"},{"key":"ref22","first-page":"5998","article-title":"Attention is all you need","volume-title":"Proc. NIPS","author":"Vaswani"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1162\/089976600300015349"},{"key":"ref24","first-page":"4696","article-title":"When does label smoothing help?","volume-title":"Proc. NIPS","author":"M\u00fcller"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/FG.2019.8756582"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414878"}],"event":{"name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Rhodes Island, Greece","start":{"date-parts":[[2023,6,4]]},"end":{"date-parts":[[2023,6,10]]}},"container-title":["ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10094559\/10094560\/10096858.pdf?arnumber=10096858","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,11]],"date-time":"2024-01-11T20:01:06Z","timestamp":1705003266000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10096858\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,4]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1109\/icassp49357.2023.10096858","relation":{},"subject":[],"published":{"date-parts":[[2023,6,4]]}}}