125
click to vote
ICASSP
2009
IEEE
15 years 10 months ago
2009
IEEE
We derive an efficient learning algorithm for model-based source separation for use on single channel speech mixtures where the precise source characteristics are not known a pri...