{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:51:36Z","timestamp":1764399096738,"version":"3.46.0"},"reference-count":30,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"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":[[2025,10,22]]},"DOI":"10.1109\/apsipaasc65261.2025.11249350","type":"proceedings-article","created":{"date-parts":[[2025,11,28]],"date-time":"2025-11-28T18:40:26Z","timestamp":1764355226000},"page":"561-566","source":"Crossref","is-referenced-by-count":0,"title":["Constructing an In-the-Wild Spoken Dialogue Dataset Based on Youtube Dialogue Videos"],"prefix":"10.1109","author":[{"given":"Yuki","family":"Sato","sequence":"first","affiliation":[{"name":"Nagoya University,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sanae","family":"Yamashita","sequence":"additional","affiliation":[{"name":"Nagoya University,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shinnosuke","family":"Takamichi","sequence":"additional","affiliation":[{"name":"Keio University,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ryuichiro","family":"Higashinaka","sequence":"additional","affiliation":[{"name":"Nagoya University,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-32967-3"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W17-5527"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10955"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/0378-2166(95)00036-4"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICSDA.2011.6086001"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-806"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"ref8","article-title":"The Fisher corpus: A resource for the next generations of speech-to-text","volume-title":"Proceedings of the 4th International Conference on Language Resources and Evaluation","author":"Cieri","year":"2004"},{"volume-title":"CALLHOME Japanese Speech, LDC96S37, Linguistic Data Consortium","year":"1996","author":"Barbara","key":"ref9"},{"key":"ref10","first-page":"5587","article-title":"Design and evaluation of the corpus of everyday Japanese conversation","volume-title":"Proceedings of the 13th Language Resources and Evaluation Conference","author":"Koiso","year":"2022"},{"key":"ref11","article-title":"Moshi: A speech-text foundation model for real-time dialogue","author":"D\u00e9fossez","year":"2024","journal-title":"arXiv preprint"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1192"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.709"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2025-2564"},{"key":"ref15","article-title":"Common Voice: A massively-multilingual speech corpus","author":"Ardila","year":"2019","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1965"},{"key":"ref17","article-title":"J-CHAT: Japanese large-scale spoken dialogue corpus for spoken dialogue language modeling","author":"Nakata","year":"2024","journal-title":"arXiv preprint"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747325"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/11939993_73"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3675166"},{"key":"ref21","first-page":"615","article-title":"\u201cAlexa in the wild\u201d - collecting unconstrained conversations with a modern voice assistant in a public environment","volume-title":"Proceedings of the 12th Language Resources and Evaluation Conference","author":"Siegert","year":"2020"},{"key":"ref22","article-title":"JTubeSpeech: Corpus of Japanese speech collected from YouTube for speech recognition and speaker verification","author":"Takamichi","year":"2021","journal-title":"arXiv preprint"},{"key":"ref23","first-page":"1739","article-title":"Diversity, density, and homogeneity: Quantitative characteristic metrics for text collections","volume-title":"Proceedings of the Twelfth Language Resources and Evaluation Conference","author":"Lai","year":"2020"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095161"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/SSW.2016-30"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-299"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414878"},{"key":"ref28","first-page":"559","article-title":"MuLan: A joint embedding of music audio and natural language","volume-title":"Proceedings of the 23rd International Siciety for Music Information Retrieval Conference","author":"Huang","year":"2022"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389693"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-205"}],"event":{"name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","start":{"date-parts":[[2025,10,22]]},"location":"Singapore, Singapore","end":{"date-parts":[[2025,10,24]]}},"container-title":["2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11248853\/11248968\/11249350.pdf?arnumber=11249350","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:51:00Z","timestamp":1764399060000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11249350\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":30,"URL":"https:\/\/doi.org\/10.1109\/apsipaasc65261.2025.11249350","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}