A noise robust algorithm for underdetermined source separation

We present a method for source separation of speech and music signals when the number of sources is larger than the number of observation sensors, a problem known as underdetermined source separation. The method uses an iterative expectation-maximization procedure to estimate demixing parameters including frequency-dependent attenuation and delay. To deal with noise distortion, the method treats noise explicitly as one of its parameters but identifies sources implicitly using a posteriori probabilities. We also extend the method to incorporate prior source statistics, represented as Gaussian mixture model. We evaluated the method in a set of noise conditions, and observed significant and consistent performance improvements than alternative methods.