Speech separation by efficient combinatorial decoding of speech mixtures.

Manuel Reyes-Gomez Nebojsa Jojic

Published in: ICME (2009)

Keyphrases

speech recognition
speech signal
text to speech
audio visual
neural network
cost effective
multi stream
finite state transducers
computationally expensive
multimedia
mixture model
computationally efficient
broadcast news
lexical features
fundamental frequency
endpoint detection