An RNN-Based Quantized F0 Model with Multi-Tier Feedback Links for Text-to-Speech Synthesis.

Xin Wang, Shinji Takaki, Junichi Yamagishi

Browse the full Interspeech paper archive.