A Style Control Technique for HMM-Based Expressive Speech Synthesis

Takashi NOSE  Junichi YAMAGISHI  Takashi MASUKO  Takao KOBAYASHI  

IEICE TRANSACTIONS on Information and Systems   Vol.E90-D   No.9   pp.1406-1413
Publication Date: 2007/09/01
Online ISSN: 1745-1361
DOI: 10.1093/ietisy/e90-d.9.1406
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Speech and Hearing
HMM-based speech synthesis,  speaking style,  emotional expression,  style interpolation,  hidden semi-Markov model (HSMM),  multiple-regression HSMM (MRHSMM),  

Full Text: PDF(914.1KB)>>
Buy this Article

This paper describes a technique for controlling the degree of expressivity of a desired emotional expression and/or speaking style of synthesized speech in an HMM-based speech synthesis framework. With this technique, multiple emotional expressions and speaking styles of speech are modeled in a single model by using a multiple-regression hidden semi-Markov model (MRHSMM). A set of control parameters, called the style vector, is defined, and each speech synthesis unit is modeled by using the MRHSMM, in which mean parameters of the state output and duration distributions are expressed by multiple-regression of the style vector. In the synthesis stage, the mean parameters of the synthesis units are modified by transforming an arbitrarily given style vector that corresponds to a point in a low-dimensional space, called style space, each of whose coordinates represents a certain specific speaking style or emotion of speech. The results of subjective evaluation tests show that style and its intensity can be controlled by changing the style vector.