{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,10,29]],"date-time":"2024-10-29T19:34:56Z","timestamp":1730230496459,"version":"3.28.0"},"reference-count":22,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"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":[[2022,5,23]]},"DOI":"10.1109\/icassp43922.2022.9747050","type":"proceedings-article","created":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T19:50:34Z","timestamp":1651089034000},"page":"7477-7481","source":"Crossref","is-referenced-by-count":1,"title":["Varianceflow: High-Quality and Controllable Text-to-Speech using Variance Information via Normalizing Flow"],"prefix":"10.1109","author":[{"given":"Yoonhyung","family":"Lee","sequence":"first","affiliation":[{"name":"Seoul National University,Dept. of Electrical and Computer Engineering,Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinhyeok","family":"Yang","sequence":"additional","affiliation":[{"name":"Speech AI Lab, NCSOFT,Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kyomin","family":"Jung","sequence":"additional","affiliation":[{"name":"Seoul National University,Dept. of Electrical and Computer Engineering,Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-866"},{"article-title":"Density estimation using real NVP","year":"0","author":"dinh","key":"ref11"},{"key":"ref12","first-page":"2784","article-title":"Network-to-network translation with conditional invertible neural networks","volume":"33","author":"rombach","year":"2020","journal-title":"Proc Advances in Neural Information Processing Systems"},{"key":"ref13","article-title":"NICE: non-linear independent components estimation","author":"dinh","year":"2015","journal-title":"3rd Int Conf on Learning Representations ICLR 2015 San Diego CA USA May 7-9 2015 Workshop Track Proceedings"},{"key":"ref14","article-title":"Neural spline flows","volume":"32","author":"durkan","year":"2019","journal-title":"Proc Advances in Neural Information Processing Systems"},{"article-title":"The lj speech dataset","year":"2017","author":"ito","key":"ref15"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.wocn.2018.07.001"},{"key":"ref17","first-page":"17022","article-title":"Hifigan: Generative adversarial networks for efficient and high fidelity speech synthesis","volume":"33","author":"kong","year":"2020","journal-title":"Proc Advances in Neural Information Processing Systems"},{"key":"ref18","first-page":"5530","article-title":"Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech","volume":"139","author":"kim","year":"2021","journal-title":"Proc of the International Conference on Machine Learning (ICML)"},{"article-title":"Decoupled weight decay regularization","year":"0","author":"loshchilov","key":"ref19"},{"key":"ref4","first-page":"8599","article-title":"Grad-tts: A diffusion probabilistic model for text-to-speech","volume":"139","author":"popov","year":"2021","journal-title":"Proc Int Conf on Machine Learning (ICML)"},{"key":"ref3","first-page":"8067","article-title":"Glow-tts: A generative flow for text-to-speech via monotonic alignment search","volume":"33","author":"kim","year":"2020","journal-title":"Proc Advances in Neural Information Processing Systems"},{"key":"ref6","article-title":"Fastspeech: Fast, robust and controllable text to speech","volume":"32","author":"ren","year":"2019","journal-title":"Proc Advances in Neural Information Processing Systems"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-971"},{"article-title":"Fastspeech 2: Fast and high-quality end-to-end text to speech","year":"0","author":"ren","key":"ref8"},{"key":"ref7","first-page":"10192","article-title":"Non-autoregressive neural text-to-speech","author":"peng","year":"2020","journal-title":"Proc Int Conf on Machine Learning (ICML)"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33016706"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461368"},{"key":"ref9","first-page":"6588","article-title":"Fastpitch: Parallel text-to-speech with pitch prediction","author":"la?cucki","year":"2021","journal-title":"Proc ICASSP"},{"key":"ref20","article-title":"Attention is all you need","volume":"30","author":"vaswani","year":"2017","journal-title":"Proc Advances in Neural Information Processing Systems"},{"key":"ref22","article-title":"Flowtron: an autoregressive flow-based generative network for text-to-speech synthesis","author":"valle","year":"2021","journal-title":"Proc Int Conf on Learning Representations (ICLR)"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2009.4960497"}],"event":{"name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2022,5,23]]},"location":"Singapore, Singapore","end":{"date-parts":[[2022,5,27]]}},"container-title":["ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9745891\/9746004\/09747050.pdf?arnumber=9747050","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,22]],"date-time":"2022-08-22T20:14:45Z","timestamp":1661199285000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9747050\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,23]]},"references-count":22,"URL":"https:\/\/doi.org\/10.1109\/icassp43922.2022.9747050","relation":{},"subject":[],"published":{"date-parts":[[2022,5,23]]}}}