The NICT Entry for the Blizzard Challenge 2009: an Enhanced HMM-based Speech Synthesis System with Trajectory Training Considering Global Variance and State-Dependent Mixed Excitation


NAIST \| Library	Advanced Search	Japanese \| English

Please use this identifier to cite or link to this item: http://hdl.handle.net/10061/7927

Title:	The NICT Entry for the Blizzard Challenge 2009: an Enhanced HMM-based Speech Synthesis System with Trajectory Training Considering Global Variance and State-Dependent Mixed Excitation
Authors:	Ranniery Maia Tomoki Toda Shinsuke Sakai Yoshinori Shiga Jinfu Ni Hisashi Kawai Keiichi Tokuda Minoru Tsuzaki Satoshi Nakamura
Keywords:	speech synthesis Blizzard Challenge HMM-based speech synthesis trajectry HMM residual modeling
Issue Date:	Sep-2009
Abstract:	This paper describes the NICT speech synthesis system submitted to the Blizzard Challenge 2009: a hidden Markov model (HMM)-based synthesizer constructed by training trajectory HMMs considering global variance. To improve naturalness of the synthesized speech a mixed excitation approach based on closed-loop residual modeling through the training of statedependent filters is employed. According to the of cial results the system in question performs well in terms of naturalness and intelligibility although synthesized speech does not sound very similar to the original speaker.
Description:	Blizzard Challenge 2009 Workshop, September 4, 2009, Edinburgh, UK.
URI:	http://hdl.handle.net/10061/7927
Text Version:	Publisher
Appears in Collections:	情報科学研究科 / Graduate School of Information Science

Files in This Item:

File	Size	Format
BCW_2009.pdf	444.5 kB	Adobe PDF	View/Open