{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,7]],"date-time":"2025-02-07T05:27:48Z","timestamp":1738906068682,"version":"3.37.0"},"reference-count":63,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2024,12,1]],"date-time":"2024-12-01T00:00:00Z","timestamp":1733011200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,12,1]],"date-time":"2024-12-01T00:00:00Z","timestamp":1733011200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,12,1]],"date-time":"2024-12-01T00:00:00Z","timestamp":1733011200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100003696","name":"Electronics and Telecommunications Research Institute","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003696","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Korean Government"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE J. Sel. Top. Signal Process."],"published-print":{"date-parts":[[2024,12]]},"DOI":"10.1109\/jstsp.2024.3437155","type":"journal-article","created":{"date-parts":[[2024,8,22]],"date-time":"2024-08-22T18:29:11Z","timestamp":1724351351000},"page":"1531-1543","source":"Crossref","is-referenced-by-count":0,"title":["Optimizations of Neural Audio Coder Toward Perceptual Transparency"],"prefix":"10.1109","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5114-5459","authenticated-orcid":false,"given":"Joon","family":"Byun","sequence":"first","affiliation":[{"name":"Computer Science Department, Yonsei University, Wonju, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8286-5182","authenticated-orcid":false,"given":"Seungmin","family":"Shin","sequence":"additional","affiliation":[{"name":"Computer Science Department, Yonsei University, Wonju, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7407-2440","authenticated-orcid":false,"given":"Seorim","family":"Hwang","sequence":"additional","affiliation":[{"name":"Computer Science Department, Yonsei University, Wonju, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3396-1137","authenticated-orcid":false,"given":"Jongmo","family":"Sung","sequence":"additional","affiliation":[{"name":"Electronics and Telecommunications Research Institute, Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6254-2062","authenticated-orcid":false,"given":"Seungkwon","family":"Beack","sequence":"additional","affiliation":[{"name":"Electronics and Telecommunications Research Institute, Daejeon, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3274-076X","authenticated-orcid":false,"given":"Youngcheol","family":"Park","sequence":"additional","affiliation":[{"name":"Software Division, Yonsei University, Wonju, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1002\/0470041978"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2008.918979"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.5555\/2969033.2969125"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1816"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2018.2887337"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096056"},{"key":"ref9","first-page":"1149","article-title":"Experimental evaluation of WaveRNN predictor for audio lossless coding","volume-title":"Proc. Int. Workshop Nonlinear Circuits, Commun. Signal Process.","author":"Amada","year":"2019"},{"article-title":"WaveNet: A generative model for raw audio","year":"2016","author":"Oord","key":"ref10"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683277"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746296"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3129994"},{"article-title":"High fidelity neural audio compression","year":"2022","author":"Dfossez","key":"ref14"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2019.2908700"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462529"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682435"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096243"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.3403\/bsenisoiec11172"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.3403\/01007714"},{"key":"ref21","article-title":"High-quality, low-delay music coding in the opus codec","volume-title":"Proc. Audio Eng. Soc. Conv.","volume":"135","author":"Valin","year":"2013"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2020.3039765"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747527"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/MP.2006.1664069"},{"key":"ref25","first-page":"6306","article-title":"Neural discrete representation learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Oord","year":"2017"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.11591\/ijece.v11i4.pp3459-3469"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/WASPAA52581.2021.9632723"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447532"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.23919\/APSIPAASC55919.2022.9980064"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3343031.3351148"},{"volume-title":"P. 862.2: Wideband Extension to Recommendation P. 862 for the Assessment of Wideband Telephone Networks and Speech Codecs","year":"2005","author":"Rec","key":"ref31"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2018.2821903"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2018.2871419"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461487"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054347"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-2151"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-955"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747575"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096009"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-2305"},{"key":"ref41","first-page":"1","article-title":"End-to-end optimized image compression","volume-title":"Proc. Int. Conf. Learn. Representation","author":"Ball","year":"2017"},{"key":"ref42","first-page":"1","article-title":"Variational image compression with a scale hyperprior","volume-title":"Proc. Int. Conf. Learn. Representation","author":"Ball","year":"2018"},{"key":"ref43","first-page":"1","article-title":"Auto-encoding variational bayes","volume-title":"Proc. Int. Conf. Learn. Representation","author":"Kingma","year":"2014"},{"issue":"14","key":"ref44","doi-asserted-by":"crossref","first-page":"1","DOI":"10.3390\/app9142854","article-title":"Psychoacoustic models for perceptual audio codinga tutorial review","volume":"9","author":"Herre","year":"2019","journal-title":"Appl. Sci."},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/5.842996"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/30.964154"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1988.197157"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096509"},{"key":"ref49","first-page":"10771","article-title":"Joint autoregressive and hierarchical priors for learned image compression","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Minnen","year":"2018"},{"key":"ref50","first-page":"1","article-title":"Context-adaptive entropy model for end-to-end optimized image compression","volume-title":"Proc. Int. Conf. Learn. Representation","author":"Lee","year":"2019"},{"key":"ref51","first-page":"1141","article-title":"Soft-to-hard vector quantization for end-to-end learning compressible representations","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Agustsson","year":"2017"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10084"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1228"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00105"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10097001"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1016"},{"key":"ref57","first-page":"27980","article-title":"High-fidelity audio compression with improved RVQGAN","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kumar","year":"2024"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10446359"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/QoMEX48832.2020.9123150"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3069302"},{"year":"2011","key":"ref61","article-title":"Unified speech and audio coding verification test report"},{"year":"2018","key":"ref62","article-title":"BBC sound effects dataset"},{"volume-title":"Method for the Subjective Assessment of Intermediate Quality Level of Audio Systems","year":"2003","key":"ref63"}],"container-title":["IEEE Journal of Selected Topics in Signal Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/4200690\/10874827\/10643673.pdf?arnumber=10643673","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,6]],"date-time":"2025-02-06T18:37:45Z","timestamp":1738867065000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10643673\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12]]},"references-count":63,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/jstsp.2024.3437155","relation":{},"ISSN":["1932-4553","1941-0484"],"issn-type":[{"type":"print","value":"1932-4553"},{"type":"electronic","value":"1941-0484"}],"subject":[],"published":{"date-parts":[[2024,12]]}}}