Prosodic vs. segmental contributions to naturalness in a diphone synthesizer.

H. Timothy Bunnell Steven R. Hoskins Debra Yarrington

Published in: ICSLP (1998)

Keyphrases

text to speech
text to speech synthesis
hidden markov models
prosodic features
speech recognition
information technology
speech synthesis
data sets
machine learning
search engine
metadata
bayesian networks
pattern recognition
pairwise
state transitions
control signals