Self-Supervised Metric Learning With Graph Clustering For Speaker Diarization

TL;DR

This paper introduces a self-supervised metric learning approach for speaker diarization that jointly learns speaker embeddings and similarity metrics, significantly improving clustering accuracy over traditional methods.

Contribution

It presents a novel joint learning framework combining self-supervised metric learning with graph clustering for speaker diarization, outperforming existing approaches.

Findings

Achieved 60% relative DER reduction on AMI dataset.

Achieved 7% relative DER reduction on DIHARD dataset.

Demonstrated effectiveness of self-supervised metric learning in speaker diarization.

Abstract

In this paper, we propose a novel algorithm for speaker diarization using metric learning for graph based clustering. The graph clustering algorithms use an adjacency matrix consisting of similarity scores. These scores are computed between speaker embeddings extracted from pairs of audio segments within the given recording. In this paper, we propose an approach that jointly learns the speaker embeddings and the similarity metric using principles of self-supervised learning. The metric learning network implements a neural model of the probabilistic linear discriminant analysis (PLDA). The self-supervision is derived from the pseudo labels obtained from a previous iteration of clustering. The entire model of representation learning and metric learning is trained with a binary cross entropy loss. By combining the self-supervision based metric learning along with the graph-based clustering algorithm, we achieve significant relative improvements of 60% and 7% over the x-vector PLDA agglomerative hierarchical clustering (AHC) approach on AMI and the DIHARD datasets respectively in terms of diarization error rates (DER).