Speaker Diarization with Overlapping Community Detection Using Graph Attention Networks and Label Propagation Algorithm

TL;DR

This paper introduces a novel graph attention network and label propagation-based method for speaker diarization that effectively handles overlapping speech segments, significantly reducing error rates in challenging datasets.

Contribution

The paper presents OCDGALP, a new framework combining graph attention networks and label propagation for improved overlapping speaker diarization.

Findings

Achieved 15.94% DER on DIHARD-III without oracle VAD.

Achieved 11.07% DER with oracle VAD.

Outperformed traditional clustering methods in overlapping speech scenarios.

Abstract

In speaker diarization, traditional clustering-based methods remain widely used in real-world applications. However, these methods struggle with the complex distribution of speaker embeddings and overlapping speech segments. To address these limitations, we propose an Overlapping Community Detection method based on Graph Attention networks and the Label Propagation Algorithm (OCDGALP). The proposed framework comprises two key components: (1) a graph attention network that refines speaker embeddings and node connections by aggregating information from neighboring nodes, and (2) a label propagation algorithm that assigns multiple community labels to each node, enabling simultaneous clustering and overlapping community detection. Experimental results show that the proposed method significantly reduces the Diarization Error Rate (DER), achieving a state-of-the-art 15.94% DER on the DIHARD-III dataset without oracle Voice Activity Detection (VAD), and an impressive 11.07% with oracle VAD.