Jointly Distributed Mean and Mixing Coefficients for Bayesian Source Separation using MCMC and ICM.

Presents the jointly distributed mean and mixing coefficients for Bayesian source separation using the Markov chain Monte Carlo method and the iterated conditional modes algorithm. Details on the multivariate Bayesian statistical approach and the linear synthesis model; Information on the conjugate estimation; Gibbs sampling estimation of the posterior.