Enhancing Hybrid Self-attention Structure with Relative-position-aware Bias for Speech Synthesis.

Shan Yang Heng Lu Shiying Kang Lei Xie Dong Yu

Published in: ICASSP (2019)

Keyphrases

speech synthesis
relative position
speech recognition
text to speech
three dimensional
neural network
image processing
low level
data mining
face recognition
pairwise
geometric properties
vocal tract