Incorporating dynamic features into minimum generation error training for HMM-based speech synthesis.

Duy Khanh Ninh Masanori Morise Yoichi Yamashita

Published in: ISCSLP (2012)

Keyphrases

speech synthesis
speech recognition
hidden markov models
text to speech
prosodic features
vocal tract
error rate
training set
training process
test set
training error
training algorithm
threshold values
generation process
training phase
generalization error
training examples
computer software
data sets
feature space