Login / Signup
Using a Pitch-Synchronous Residual Codebook for Hybrid HMM/Frame Selection Speech Synthesis.
Thomas Drugman
Alexis Moinet
Thierry Dutoit
Geoffrey Wilfart
Published in:
CoRR (2019)
Keyphrases
</>
speech synthesis
speech recognition
hidden markov models
text to speech
vector quantization
prosodic features
vocal tract
language model
video frames
automatic speech recognition
feature vectors
asynchronous communication
speech corpus
forward backward
hybrid learning
bag of words
image representation