Non-negative tensor factorisation of modulation spectrograms for monaural sound source separation.

Tom Barker Tuomas Virtanen

Published in: INTERSPEECH (2013)

Keyphrases

source separation
single channel
blind source separation
sound source
independent component analysis
multi channel
speech signal
temporal structure
denoising
audio features
frequency domain
dimensionality reduction
factor analysis
simple linear
image retrieval
gaussian process