A study on pitch pattern generation using HMM-based statistical information

This paper describes a novel pitch pattern generation method for speech synthesis using Hidden Markov Models (HMMs). In the proposed method, the F 0 contours of minor phrase are modeled by HMMs (pitch-HMMs). The pitch-HMMs are trained using F 0 and F 0 considering phonetic environments (e.g. accent type, mora count, mora position, phonemic category, etc.). To evaluate the pitch-HMMs, accent identi cation experiments are performed. The results indicate that the pitch-HMMs can capture the movement in F 0 contours appropriately. In the F 0 contour generation experiments, the proposed method yields an averaged root mean square error of 132cent (equivalent to 9.2Hz at 120Hz) between the original and the generated F 0 contours. Furthermore, an application of the proposed method to text-to-speech system is also discussed.