Variational learning and inference algorithms for extended Gaussian mixture model

In this paper, in order to properly evaluate the relative importance of priors and observed data in the Bayesian framework, we propose an extended Gaussian mixture model (EGMM) and design the corresponding learning inference algorithms. First, we define the likelihood function of the EGMM and then propose the variational learning algorithm for this EGMM. Moreover, the proposed model and approach are applied to speaker recognition. Experimental results demonstrate that this new approach generalizes the traditional GMM, offering a more powerful performance.