Constrained structural maximum a posteriori linear regression for average-voice-based speech synthesis.

Yuji Nakano, Makoto Tachibana, Junichi Yamagishi, Takao Kobayashi

Browse the full Interspeech paper archive.