Perceptual distortion analysis and quality estimation of prosodymodified speech for TD-PSOLA

TD-PSOLA is one of the most widely used prosodic modification techniques. However, perceptible distortions are introduced occasionally and how TD-PSOLA affects speech quality has not been fully understood and controlled. In this paper, we present a quality estimation method before performing modification. By exploiting relationship between prosodic modifications and subjective scores, 27 distance measures are proposed and respective performances are given and compared. Extensive search is used to find every possible combination among these measures, and the best correlation between the predicted and subjective scores is 87.6%, which can be obtained by linear regression of 4 proposed distance measures. The proposed method does not require synthesizing target and can be used both in online unit selection and off-line corpus design of TTS systems.