{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T18:28:46Z","timestamp":1776882526688,"version":"3.51.2"},"reference-count":51,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,12,16]],"date-time":"2023-12-16T00:00:00Z","timestamp":1702684800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,12,16]],"date-time":"2023-12-16T00:00:00Z","timestamp":1702684800000},"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":[[2023,12,16]]},"DOI":"10.1109\/asru57964.2023.10389618","type":"proceedings-article","created":{"date-parts":[[2024,1,19]],"date-time":"2024-01-19T18:38:40Z","timestamp":1705689520000},"page":"1-8","source":"Crossref","is-referenced-by-count":12,"title":["Scenario-Aware Audio-Visual TF-Gridnet for Target Speech Extraction"],"prefix":"10.1109","author":[{"given":"Zexu","family":"Pan","sequence":"first","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gordon","family":"Wichern","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yoshiki","family":"Masuyama","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fran\u00e7ois G.","family":"Germain","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sameer","family":"Khurana","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chiori","family":"Hori","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jonathan Le","family":"Roux","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1653"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461375"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413776"},{"key":"ref4","article-title":"Predict-and-Update network: Audio-visual speech recognition inspired by human speech perception","author":"Wang","year":"2022","journal-title":"arXiv preprint arXiv:2209.01768"},{"key":"ref5","article-title":"Is someone speaking? Exploring long-term temporal features for audio-visual active speaker detection","volume-title":"Proc. ACM Multimedia","author":"Tao"},{"issue":"1","key":"ref6","first-page":"117","article-title":"The cocktail party phenomenon: A review of research on speech intelligibility in multiple-talker conditions","volume":"86","author":"Bronkhorst","year":"2000","journal-title":"Acta Acust. United Acust"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2023.3240008"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2020.2987429"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-1101"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2020-1397"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054222"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682245"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2418"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054683"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1162\/1064546053278973"},{"key":"ref16","volume-title":"The theory of neuronal group selection","author":"Edelman","year":"1987"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0004638"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.3675-12.2013"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.1396-16.2016"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1065"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3153258"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2022.3175130"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3066303"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2018-1400"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003983"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746866"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054180"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-2545"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-11183"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/jstsp.2020.2987209"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2019.2928140"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682061"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201357"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-1513"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2019.2915167"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414023"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10097232"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054266"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3205759"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/SLT54892.2023.10023284"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-157"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/icassp49357.2023.10094992"},{"key":"ref43","article-title":"Multi-channel target speaker extraction with refinement: The WAVLab submission to the second clarity enhancement challenge","volume-title":"Proc. Clarity","author":"Cornell"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1943"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683855"},{"key":"ref46","article-title":"LRS3TED: a large-scale dataset for visual speech recognition","author":"Afouras","year":"2018","journal-title":"arXiv preprint arXiv:1809.00496"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1574"},{"key":"ref48","article-title":"DEMAND: a collection of multi-channel recordings of acoustic noise in diverse environments","volume-title":"Proc. Meetings Acoust","author":"Thiemann"},{"key":"ref49","article-title":"Deep speech enhancement challenge at ICASSP 2023","volume-title":"Proc. ICASSP","author":"Dubey"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3133208"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414878"}],"event":{"name":"2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","location":"Taipei, Taiwan","start":{"date-parts":[[2023,12,16]]},"end":{"date-parts":[[2023,12,20]]}},"container-title":["2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10388490\/10389614\/10389618.pdf?arnumber=10389618","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,23]],"date-time":"2024-01-23T16:35:24Z","timestamp":1706027724000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10389618\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,16]]},"references-count":51,"URL":"https:\/\/doi.org\/10.1109\/asru57964.2023.10389618","relation":{},"subject":[],"published":{"date-parts":[[2023,12,16]]}}}