Deep discriminant analysis for i-vector based robust speaker recognition


Linear Discriminant Analysis (LDA) has been used as a standard post-processing procedure in many state-of-the-art speaker recognition tasks. Through maximizing the inter-speaker difference and minimizing the intra-speaker variation, LDA projects i-vectors to a lower-dimensional and more discriminative subspace. In this paper, we propose a neural network based compensation scheme(termed as deep discriminant analysis, DDA) for i-vector based speaker recognition, which shares the idea with LDA. Optimized against softmax loss and center loss at the same time, the proposed method learns a more compact and discriminative embedding space. Compared with the Gaussian distribution assumption of data and the learned linear projection in LDA, the proposed method doesn’t pose any assumptions on data and can learn a non-linear projection function. Experiments are carried out on a short-duration text-independent dataset based on the SRE Corpus, noticeable performance improvement can be observed against the normal LDA or PLDA methods.

In the 11th International Symposium on Chinese Spoken Language Processing(ISCSLP), Taipei, Taiwan, China, 2018.