A HMM-based fuzzy affective model for emotional speech synthesis

Existing emotional speech synthesis applications usually distinguish between a small number of emotions in speech. However this set of so called basic emotions in speech varies from one application to another depending on their according needs. In order to support such differing application needs an emotional speech fuzzy model is presented. In addition to existing models it supports also the synthesis of derived emotions which are combinations of basic emotions in speech. We show the application of this model by a prosody based Hidden Markov Models(HMM). The approach is based on emotional speech corpus database that trained by HMM. This approach use three kinds of emotional speech corpus (anger, happiness, and sadness) from recordings of a male and a female speaker of Chinese and English. Both the selection of features and the design of the synthesis are addressed.