Automatic segmentation for czech concatenative speech synthesis using statistical approach with boundary-specific correction

Abstract

This paper deals with the problems of automatic segmentation for the purposes of Czech concatenative speech synthesis. Statistical approach to speech segmentation using hidden Markov models (HMMs) is applied in the baseline system. Several improvements of this system are then proposed to get more accurate segmentation results. These enhancements mainly concern the various strategies of HMM initialization (flat-start initialization, hand-labeled or speaker independent HMM bootstrapping). Since HTK, the hidden Markov model toolkit, was utilized in our work, a correction of the output boundary placements is proposed to reflect speech parameterization mechanism. An objective comparison of various automatic methods and manual segmentation is performed to find out the best method. The best results were obtained for boundary-specific statistical correction of the segmentation that resulted from bootstrapping with hand-labeled HMMs (96% segmentation accuracy in tolerance region 20 ms).