Smoothed N-best-based speaker adaptation for speech recognition.

Tomoko Matsui Tatsuo Matsuoka Sadaoki Furui

Published in: ICASSP (1997)

Keyphrases

speaker adaptation
speech recognition
automatic speech recognition
speaker dependent
speaker independent

speech recognizer
hidden markov models
language model
speech synthesis
speech signal

noisy environments
pattern recognition
speech recognition systems
maximum likelihood
image processing

computer vision
n gram
speaker identification
probabilistic model
feature selection