Improving Emotional Speech Synthesis by Using SUS-Constrained VAE and Text Encoder Aggregation.

Fengyu Yang, Jian Luan, Yujun Wang

Browse the full ICASSP paper archive.