{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T16:19:42Z","timestamp":1781713182292,"version":"3.54.5"},"publisher-location":"ISCA","reference-count":0,"publisher":"ISCA","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"DOI":"10.21437\/interspeech.2024-764","type":"proceedings-article","created":{"date-parts":[[2024,9,1]],"date-time":"2024-09-01T07:10:12Z","timestamp":1725174612000},"page":"42-46","source":"Crossref","is-referenced-by-count":3,"title":["AFL-Net: Integrating Audio, Facial, and Lip Modalities with a Two-step Cross-attention for Robust Speaker Diarization in the Wild"],"prefix":"10.21437","author":[{"given":"YongKang","family":"Yin","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xu","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ying","family":"Shan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"YueXian","family":"Zou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"8866","published-online":{"date-parts":[[2024,9,1]]},"event":{"name":"Interspeech 2024","acronym":"interspeech_2024"},"container-title":["Interspeech 2024"],"original-title":[],"deposited":{"date-parts":[[2024,11,8]],"date-time":"2024-11-08T20:24:43Z","timestamp":1731097483000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.isca-archive.org\/interspeech_2024\/yin24_interspeech.html"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,1]]},"references-count":0,"URL":"https:\/\/doi.org\/10.21437\/interspeech.2024-764","relation":{},"subject":[],"published":{"date-parts":[[2024,9,1]]}}}