Login / Signup
Acoustic model training based on linear transformation and MAP modification for HSMM-based speech synthesis.
Katsumi Ogata
Makoto Tachibana
Junichi Yamagishi
Takao Kobayashi
Published in:
INTERSPEECH (2006)
Keyphrases
</>
linear transformation
speech synthesis
speech recognition
text to speech
distance metric
linear transformations
supervised dimensionality reduction
vocal tract
linear model
zernike moments
metric learning
training set
multiscale
speech signal
euclidean distance
least squares
state space
hidden markov models