{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T07:21:17Z","timestamp":1764400877871,"version":"3.46.0"},"reference-count":27,"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\/apsipaasc65261.2025.11249036","type":"proceedings-article","created":{"date-parts":[[2025,11,28]],"date-time":"2025-11-28T18:40:26Z","timestamp":1764355226000},"page":"2535-2540","source":"Crossref","is-referenced-by-count":0,"title":["Explainable Disentanglement on Discrete Speech Representations for Noise-Robust ASR"],"prefix":"10.1109","author":[{"given":"Shreyas","family":"Gopal","sequence":"first","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ashutosh","family":"Anshul","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haoyang","family":"Li","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yue Heng","family":"Yeo","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hexin","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Eng Siong","family":"Chng","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"27 980","article-title":"High-fidelity audio compression with improved rvqgan","volume-title":"Proc. NeurIPS","volume":"36","author":"Kumar","year":"2023"},{"journal-title":"High fidelity neural audio compression","year":"2022","author":"D\u00e9fossez","key":"ref2"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3129994"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-330"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10446998"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.787"},{"journal-title":"Cosyvoice 2: Scalable streaming speech synthesis with large language models","year":"2024","author":"Du","key":"ref7"},{"key":"ref8","article-title":"Moshi: A speech-text foundation model for real-time dialogue","author":"D\u00e9fossez","year":"2024","journal-title":"Kyutai, Tech. Rep."},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2024-1533"},{"key":"ref11","first-page":"28 492","article-title":"Robust speech recognition via large-scale weak supervision","volume-title":"Proc. ICML","author":"Radford","year":"2023"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-2193"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TASLPRO.2025.3566210"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2025-1292"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746929"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.23919\/APSIPAASC55919.2022.9980261"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096435"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10888065"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096603"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-964"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19772-7_29"},{"key":"ref22","article-title":"Learning latent subspaces in variational autoencoders","volume-title":"Proc. NeurIPS","volume":"31","author":"Klys","year":"2018"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-020-05270-2"},{"key":"ref24","article-title":"Neural discrete representation learning","volume-title":"Proc. NeurIPS","volume":"30","author":"Van Den Oord","year":"2017"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/ssw.2016-24"},{"key":"ref26","article-title":"Large language models are efficient learners of noise-robust speech recognition","volume-title":"Proc. ICML","author":"Hu","year":"2024"},{"volume-title":"The 4th chime speech separation and recognition challenge","year":"2018","author":"Vincent","key":"ref27"}],"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\/11249036.pdf?arnumber=11249036","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T07:17:27Z","timestamp":1764400647000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11249036\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":27,"URL":"https:\/\/doi.org\/10.1109\/apsipaasc65261.2025.11249036","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}