Presentation 2014-01-23
A Study on Hyperparameter Optimization for Speech Synthesis Based on Gaussian Process Regression
Tomoki KORIYAMA, Takashi NOSE, Takao KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In a statistical parametric speech synthesis framework based on Gaussian process regression, it is important to use an appropriate kernel function. However the parameters of the kernel function, which are hyperparameters of Gaussian processes, were not optimized in our previous work. In this study, we examine hyperparameter optimization algorithm based on an empirical Bayes approach. We show that the proposed method can enhance the predictive likelihood and improve the naturalness of synthesic speech through objective and subjective evaluation results.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) statistical parametric speech synthesis / Gaussian process / hyperparameter / kernel function
Paper # SP2013-99
Date of Issue

Conference Information
Committee SP
Conference Date 2014/1/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study on Hyperparameter Optimization for Speech Synthesis Based on Gaussian Process Regression
Sub Title (in English)
Keyword(1) statistical parametric speech synthesis
Keyword(2) Gaussian process
Keyword(3) hyperparameter
Keyword(4) kernel function
1st Author's Name Tomoki KORIYAMA
1st Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology()
2nd Author's Name Takashi NOSE
2nd Author's Affiliation School of Engineering, Tohoku University
3rd Author's Name Takao KOBAYASHI
3rd Author's Affiliation Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology
Date 2014-01-23
Paper # SP2013-99
Volume (vol) vol.113
Number (no) 404
Page pp.pp.-
#Pages 6
Date of Issue