Gmm Based on Local Robust Pca for Speaker Identification

ABSTRACT: To solve the problems of outliers and high dimensionality of training feature vectors in speaker identification, in this paper, we propose an efficient GMM based on local robust PCA with VQ. The proposed method firstly partitions the data space into several disjoint regions by VQ, and then performs robust PCA using the iteratively reweighted covariance matrix in each region. Finally, the GMM for speaker is obtained from the transformed feature vectors with reduced dimension in each region. Compared to the conventional GMM with diagonal covariance matrix, under the same performance, the proposed method needs less storage and shows faster result. Moreover, the proposed method is more robust when outliers exist.