Unit selection using pitch synchronous cross correlation for Japanese concatenative speech synthesis.

Nobuo Nukaga Ryota Kamoshida Kenji Nagamatsu

Published in: SSW (2004)

Keyphrases

cross correlation
speech synthesis
speech recognition
frequency domain
template matching
mutual information
text to speech
vocal tract
prosodic features
acoustic features
weighted distance
multiresolution
feature extraction
selection algorithm
asynchronous communication