DNN-based stochastic postfilter for HMM-based speech synthesis

Ling-Hui Chen; Tuomo Raitio; Cassia Valentini-Botinhao; Junichi Yamagishi; Zhen-Hua Ling

DNN-based stochastic postfilter for HMM-based speech synthesis

Ling-Hui Chen, Tuomo Raitio, Cassia Valentini-Botinhao, Junichi Yamagishi, Zhen-Hua Ling

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we propose a deep neural network to model the conditional probability of the spectral differences between natural and synthetic speech. This allows us to reconstruct the spectral fine structures in speech generated by HMMs. We compared the new stochastic data-driven postfilter with global variance based parameter generation and modulation spectrum enhancement. Our results confirm that the proposed method significantly improves the segmental quality of synthetic speech compared to the conventional methods.

Original language	English
Title of host publication	Interspeech 2014
Place of Publication	Singapore
Publisher	International Speech Communication Association
Pages	1954-1958
Number of pages	5
Publication status	Published - Sept 2014

Access to Document

postfilter_IS14Final published version, 522 KB

http://www.isca-speech.org/archive/interspeech_2014/i14_1954.html

Simple4All: Speech synthesis that improves through adaptive learning
King, S. & Renals, S.
EU government bodies
1/11/11 → 31/10/14
Project: Research

File

Cite this

@inproceedings{24ab5cc8f3274bcc91cec74c742e0010,

title = "DNN-based stochastic postfilter for HMM-based speech synthesis",

abstract = "In this paper we propose a deep neural network to model the conditional probability of the spectral differences between natural and synthetic speech. This allows us to reconstruct the spectral fine structures in speech generated by HMMs. We compared the new stochastic data-driven postfilter with global variance based parameter generation and modulation spectrum enhancement. Our results confirm that the proposed method significantly improves the segmental quality of synthetic speech compared to the conventional methods.",

author = "Ling-Hui Chen and Tuomo Raitio and Cassia Valentini-Botinhao and Junichi Yamagishi and Zhen-Hua Ling",

year = "2014",

month = sep,

language = "English",

pages = "1954--1958",

booktitle = "Interspeech 2014",

publisher = "International Speech Communication Association",

}

TY - GEN

T1 - DNN-based stochastic postfilter for HMM-based speech synthesis

AU - Chen, Ling-Hui

AU - Raitio, Tuomo

AU - Valentini-Botinhao, Cassia

AU - Yamagishi, Junichi

AU - Ling, Zhen-Hua

PY - 2014/9

Y1 - 2014/9

N2 - In this paper we propose a deep neural network to model the conditional probability of the spectral differences between natural and synthetic speech. This allows us to reconstruct the spectral fine structures in speech generated by HMMs. We compared the new stochastic data-driven postfilter with global variance based parameter generation and modulation spectrum enhancement. Our results confirm that the proposed method significantly improves the segmental quality of synthetic speech compared to the conventional methods.

AB - In this paper we propose a deep neural network to model the conditional probability of the spectral differences between natural and synthetic speech. This allows us to reconstruct the spectral fine structures in speech generated by HMMs. We compared the new stochastic data-driven postfilter with global variance based parameter generation and modulation spectrum enhancement. Our results confirm that the proposed method significantly improves the segmental quality of synthetic speech compared to the conventional methods.

M3 - Conference contribution

SP - 1954

EP - 1958

BT - Interspeech 2014

PB - International Speech Communication Association

CY - Singapore

ER -

DNN-based stochastic postfilter for HMM-based speech synthesis

Abstract

Access to Document

Fingerprint

Projects

Simple4All: Speech synthesis that improves through adaptive learning

Cite this