End-to-end speaker segmentation for overlap-aware resegmentation

TL;DR

This paper introduces an end-to-end neural model for speaker segmentation that directly detects speaker turns and overlaps at high temporal resolution, improving diarization accuracy across multiple datasets.

Contribution

It presents a novel end-to-end approach for speaker segmentation that handles overlap detection and resegmentation in a unified model, outperforming traditional methods.

Findings

Achieves up to 17% reduction in diarization error rate on AMI dataset.

Effectively detects overlapped speech regions as a post-processing step.

Operates on short audio chunks with high temporal resolution.

Abstract

Speaker segmentation consists in partitioning a conversation between one or more speakers into speaker turns. Usually addressed as the late combination of three sub-tasks (voice activity detection, speaker change detection, and overlapped speech detection), we propose to train an end-to-end segmentation model that does it directly. Inspired by the original end-to-end neural speaker diarization approach (EEND), the task is modeled as a multi-label classification problem using permutation-invariant training. The main difference is that our model operates on short audio chunks (5 seconds) but at a much higher temporal resolution (every 16ms). Experiments on multiple speaker diarization datasets conclude that our model can be used with great success on both voice activity detection and overlapped speech detection. Our proposed model can also be used as a post-processing step, to detect and correctly assign overlapped speech regions. Relative diarization error rate improvement over the best considered baseline (VBx) reaches 17% on AMI, 13% on DIHARD 3, and 13% on VoxConverse.