Interference Reduction in Reverberant Speech Separation With Visual Voice Activity Detection.

Qingju Liu Andrew J. Aubrey Wenwu Wang

Published in: IEEE Trans. Multim. (2014)

Keyphrases

voice activity detection
speech signal
noisy environments
speech recognition
visual information
sound source
vocal tract
hidden markov models
visual features
non stationary
low level
unsupervised learning
neural network
facial expressions
visual cues
speaker identification
acoustic features