Speaker Identification Using GMM with Embedded AANN

Abstract

This paper proposes a modified Gaussian Mixed Model (GMM) with an embedded Auto-Associate Neural Network (AANN). It integrates the merits of GMM and AANN. GMM and AANN are trained as a whole by means of maximum likelihood. In the process of training, the parameter of GMM and AANN are updated alternately. AANN reshapes the distribution of the data and improves the similarity of the data in one class. Experiments show that the proposed system improves accuracy rate against baseline GMM at all SNR, maximum to 19%.