{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T01:40:00Z","timestamp":1740102000519,"version":"3.37.3"},"reference-count":22,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,6,27]],"date-time":"2023-06-27T00:00:00Z","timestamp":1687824000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,6,27]],"date-time":"2023-06-27T00:00:00Z","timestamp":1687824000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100016047","name":"Science Fund of the Republic of Serbia","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100016047","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007065","name":"Nvidia","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007065","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,6,27]]},"DOI":"10.1109\/iwssip58668.2023.10180259","type":"proceedings-article","created":{"date-parts":[[2023,7,18]],"date-time":"2023-07-18T17:29:52Z","timestamp":1689701392000},"page":"1-5","source":"Crossref","is-referenced-by-count":1,"title":["Cross-lingual Text-to-Speech with Prosody Embedding"],"prefix":"10.1109","author":[{"given":"Tijana","family":"Nosek","sequence":"first","affiliation":[{"name":"University of Novi Sad,Faculty of Technical Sciences,Novi Sad,Serbia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sini\u0161a","family":"Suzi\u0107","sequence":"additional","affiliation":[{"name":"University of Novi Sad,Faculty of Technical Sciences,Novi Sad,Serbia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vlado","family":"Deli\u0107","sequence":"additional","affiliation":[{"name":"University of Novi Sad,Faculty of Technical Sciences,Novi Sad,Serbia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Milan","family":"Se\u010dujski","sequence":"additional","affiliation":[{"name":"University of Novi Sad,Faculty of Technical Sciences,Novi Sad,Serbia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2668"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461368"},{"key":"ref15","article-title":"Cross-Lingual Text-to-Speech Using Multi-Task Learning and Speaker Classifier Joint Training","author":"yang","year":"2022","journal-title":"arXiv preprint arXiv 2201 08124"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1632"},{"year":"0","key":"ref20"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472737"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1587\/transinf.2015EDP7457"},{"key":"ref10","first-page":"322","article-title":"Personalized, crosslingual TTS using phonetic posteriorgrams","author":"sun","year":"2016","journal-title":"Proceedings of the 17th Annual Conference of the International Speech Communication Association INTERSPEECH 2016"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/SSW.2016-33"},{"key":"ref2","first-page":"264","article-title":"Microsoft Mulan – a bilingual TTS system","volume":"i","author":"chu","year":"2003","journal-title":"Proceedings of the IEEE International Conference on Acoustics Speech and Signal Processing ICASSP 2003"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.21437\/Eurospeech.1999-203"},{"key":"ref17","article-title":"Fastpitch: Parallel text-to-speech with pitch prediction","author":"?a?cucki","year":"2021","journal-title":"ICASSP 2021-2021 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747481"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.9781\/ijimai.2021.11.005"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096929"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2009-192"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2009.2015708"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2012.2187195"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2004-364"},{"key":"ref3","first-page":"177","article-title":"Foreign-language speech synthesis","author":"campbell","year":"1998","journal-title":"Proceedings of the 3rd ESCA\/COCOSDA Workshop (ETRW) on Speech Synthesis"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2011.5947509"},{"key":"ref5","first-page":"217","article-title":"Language independent phoneme mapping for foreign TTS","author":"badino","year":"2004","journal-title":"Proceedings of the 5th ISCA Workshop on Speech Synthesis"}],"event":{"name":"2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP)","start":{"date-parts":[[2023,6,27]]},"location":"Ohrid, North Macedonia","end":{"date-parts":[[2023,6,29]]}},"container-title":["2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10180201\/10180228\/10180259.pdf?arnumber=10180259","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,7]],"date-time":"2023-08-07T17:44:45Z","timestamp":1691430285000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10180259\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,27]]},"references-count":22,"URL":"https:\/\/doi.org\/10.1109\/iwssip58668.2023.10180259","relation":{},"subject":[],"published":{"date-parts":[[2023,6,27]]}}}