Patnet : A Phoneme-Level Autoregressive Transformer Network for Speech Synthesis.

Shiming Wang Zhenhua Ling Ruibo Fu Jiangyan Yi Jianhua Tao

Published in: ICASSP (2021)

Keyphrases

speech synthesis
autoregressive
speech recognition
text to speech
prosodic features
non stationary
moving average
random fields
gaussian markov random field
vocal tract
distribution network
random field models
information retrieval
fault diagnosis
autoregressive model
noisy environments
hidden markov models