{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:56:34Z","timestamp":1764399394452,"version":"3.46.0"},"reference-count":53,"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\/apsipaasc65261.2025.11249061","type":"proceedings-article","created":{"date-parts":[[2025,11,28]],"date-time":"2025-11-28T18:40:26Z","timestamp":1764355226000},"page":"2582-2586","source":"Crossref","is-referenced-by-count":0,"title":["From Evaluation to Optimization: Neural Speech Assessment for Downstream Applications"],"prefix":"10.1109","author":[{"given":"Yu","family":"Tsao","sequence":"first","affiliation":[{"name":"Research Center for Information Technology Innovation,Academia Sinica,Taipei,Taiwan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1250\/ast.e24.12"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414878"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3205757"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3317574"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1508"},{"journal-title":"Proc. ICLR 2024","article-title":"Self-supervised speech quality estimation and enhancement using only clean speech","author":"Fu","key":"ref6"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1802"},{"journal-title":"Proc. APSIPA ASC 2020","article-title":"STOINet: A deep learning based non-intrusive speech intelligibility assessment model","author":"Zezario","key":"ref8"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095710"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2019-2003"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413877"},{"journal-title":"Proc. NeurIPS 2021","article-title":"NORESQA: A framework for speech quality assessment using non-matching references","author":"Manocha","key":"ref12"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-407"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747222"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746395"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-439"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10922"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389788"},{"key":"ref19","article-title":"TTSDS2: Resources and benchmark for evaluating human-quality text to speech systems","author":"Minixhofer","year":"2025","journal-title":"arXiv preprint"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-970"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389763"},{"journal-title":"Proc. SLT 2024","article-title":"The VoiceMOS Challenge 2024: Beyond speech quality prediction","author":"Huang","key":"ref22"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1239"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/SLT54892.2023.10023284"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.1150"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2025-1992"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2022.12.195"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1097\/AUD.0000000000001635"},{"journal-title":"Proc. ICML 2019","article-title":"MetricGAN: Generative adversarial networks based black-box metric scores optimization for speech enhancement","author":"Fu","key":"ref29"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1844"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-936"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/icassp43922.2022.9747180"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3165442"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/MLSP55844.2023.10285913"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3328282"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2024.3426996"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/SLT61566.2024.10832332"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10890696"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2025.107562"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2025.3599453"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2425"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.23919\/EUSIPCO54536.2021.9616163"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2001.941023"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCChina.2012.6356906"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-599"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-517"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.23919\/EUSIPCO55093.2022.9909682"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2023.3263788"},{"journal-title":"Proc. ICML 2023","article-title":"MetricGAN- OKD: multi-metric optimization of MetricGAN via online knowledge distillation for speech enhancement","author":"Shin","key":"ref49"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095906"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10890542"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2011.2114881"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1145\/3422622"}],"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\/11249061.pdf?arnumber=11249061","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:53:09Z","timestamp":1764399189000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11249061\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":53,"URL":"https:\/\/doi.org\/10.1109\/apsipaasc65261.2025.11249061","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}