Unsupervised methods for speaker diarization

Abstract

Given a stream of unlabeled audio data, speaker diarization is the process of determining "who spoke when." We propose a novel approach to solving this problem by taking advantage of the effectiveness of factor analysis as a front-end for extracting speaker-specific features and exploiting the inherent variabilities in the data through the use of unsupervised methods. Upon initial evaluation, our system achieves state-of-the art results of 0.9% Diarization Error Rate in the diarization of two-speaker telephone conversations. The approach is then generalized to the problem of K-speaker diarization, for which we take measures to address issues of data sparsity and experiment with the use of the von Mises-Fisher distribution for clustering on a unit hypersphere. Our extended system performs competitively on the diarization of conversations involving two or more speakers. Finally, we present promising initial results obtained from applying variational inference on our front-end speaker representation to estimate the unknown number of speakers in a given utterance.