Presentation | 2019-06-13 A study on style transplantation modeling techniques for DNN-based speech synthesis Yoshiki Hiruta, Tomoki Koriyama, Yuuki Tachioka, Takao Kobayashi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper investigates style transplantation modeling techniques for DNN-based statistical parametric speech synthesis. The problem treated here is to generate expressive speech of a given target speaker with only using a small amount of his/her reading style speech data. For this purpose, we propose two models, which utilize i-vector as an input to DNN. The basic idea of the proposed modeling framework is to construct an acoustic model which can control voice characteristics and emotional expression and/or speaking style using multi speaker’s expressive speech data. In the proposed framework, the i-vector is used to control voice characteristics of the synthetic speech. Through objective and subjective evaluation experiments we compare the performance of the proposed style trans- plantation modeling techniques with the conventional ones in which a speaker code represented by one-hot vector is used as an input to DNN. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | DNN-based speech synthesis / style transplant / i-vector |
Paper # | SP2019-1 |
Date of Issue | 2019-06-06 (SP) |
Conference Information | |
Committee | SP |
---|---|
Conference Date | 2019/6/13(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Tokyo Institute of Technology |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Hisashi Kawai(NICT) |
Vice Chair | Akinobu Ri(Nagoya Inst. of Tech.) |
Secretary | Akinobu Ri(Kyoto Univ.) |
Assistant | Tomoki Koriyama(UTokyo) / Yusuke Ijima(NTT) |
Paper Information | |
Registration To | Technical Committee on Speech |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A study on style transplantation modeling techniques for DNN-based speech synthesis |
Sub Title (in English) | |
Keyword(1) | DNN-based speech synthesis |
Keyword(2) | style transplant |
Keyword(3) | i-vector |
1st Author's Name | Yoshiki Hiruta |
1st Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
2nd Author's Name | Tomoki Koriyama |
2nd Author's Affiliation | The University of Tokyo(The Univ. of Tokyo) |
3rd Author's Name | Yuuki Tachioka |
3rd Author's Affiliation | Denso IT Laboratory(Denso IT Lab) |
4th Author's Name | Takao Kobayashi |
4th Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
Date | 2019-06-13 |
Paper # | SP2019-1 |
Volume (vol) | vol.119 |
Number (no) | SP-80 |
Page | pp.pp.1-6(SP), |
#Pages | 6 |
Date of Issue | 2019-06-06 (SP) |