Speaker Clustering Using Dominant Sets

TL;DR

This paper introduces a novel speaker clustering method using Dominant Sets, demonstrating state-of-the-art results on the TIMIT dataset through extensive experiments and parameter stability analysis.

Contribution

The paper applies Dominant Sets, a graph-based clustering algorithm, to speaker clustering for the first time, achieving superior performance on TIMIT.

Findings

State-of-the-art clustering performance on TIMIT.

Stable results across different parameter settings.

Effective use of deep neural network features.

Abstract

Speaker clustering is the task of forming speaker-specific groups based on a set of utterances. In this paper, we address this task by using Dominant Sets (DS). DS is a graph-based clustering algorithm with interesting properties that fits well to our problem and has never been applied before to speaker clustering. We report on a comprehensive set of experiments on the TIMIT dataset against standard clustering techniques and specific speaker clustering methods. Moreover, we compare performances under different features by using ones learned via deep neural network directly on TIMIT and other ones extracted from a pre-trained VGGVox net. To asses the stability, we perform a sensitivity analysis on the free parameters of our method, showing that performance is stable under parameter changes. The extensive experimentation carried out confirms the validity of the proposed method, reporting state-of-the-art results under three different standard metrics. We also report reference baseline results for speaker clustering on the entire TIMIT dataset for the first time.