Login / Signup
Unsupervised speaker adaptation using high confidence portion recognition results by multiple recognition systems.
Tomohiro Watanabe
Hiromitsu Nishizaki
Takehito Utsuro
Seiichi Nakagawa
Published in:
INTERSPEECH (2004)
Keyphrases
</>
high confidence
pattern recognition
feature extraction
unsupervised learning
data sets
feature selection
semi supervised
maximum likelihood
character recognition