Fifth ISCA ITRW on Speech Synthesis

June 14-16, 2004
Pittsburgh, PA, USA

Multi-Source Based Acoustic Model for Speech Synthesis

Jianhua Tao, Yongguo Kang

National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciencesm, Beijing, China

Traditional source-filter model has obvious limitation for speech synthesis in pitch modification due to the lack of spectrum distortion processing. To solve the problem, the paper analyzes the spectrum features of voice source in various F0 ranges and timbres in detail, and generates Muliti-Source (MS) based on analysis results by classifying the voice source into different types. The model enhances the quality of speech synthesis in various speaking mood.

Full Paper

Bibliographic reference.  Tao, Jianhua / Kang, Yongguo (2004): "Multi-source based acoustic model for speech synthesis", In SSW5-2004, 167-172.