An improved minimum generation error based model adaptation for HMM-based speech synthesis.

Yi-Jian Wu, Long Qin, Keiichi Tokuda

Browse the full Interspeech paper archive.