End-to-End Speaker Diarization as Post-Processing

TL;DR

This paper proposes a hybrid speaker diarization approach combining clustering and end-to-end methods, significantly improving performance on multiple datasets by effectively handling overlapping speech.

Contribution

It introduces a novel iterative post-processing technique that enhances clustering-based diarization with a two-speaker end-to-end model, addressing limitations in overlapping speech detection.

Findings

Improved diarization accuracy across CALLHOME, AMI, and DIHARD II datasets.

Effective handling of overlapping speech through iterative refinement.

Consistent performance gains over state-of-the-art methods.

Abstract

This paper investigates the utilization of an end-to-end diarization model as post-processing of conventional clustering-based diarization. Clustering-based diarization methods partition frames into clusters of the number of speakers; thus, they typically cannot handle overlapping speech because each frame is assigned to one speaker. On the other hand, some end-to-end diarization methods can handle overlapping speech by treating the problem as multi-label classification. Although some methods can treat a flexible number of speakers, they do not perform well when the number of speakers is large. To compensate for each other's weakness, we propose to use a two-speaker end-to-end diarization method as post-processing of the results obtained by a clustering-based method. We iteratively select two speakers from the results and update the results of the two speakers to improve the overlapped region. Experimental results show that the proposed algorithm consistently improved the performance of the state-of-the-art methods across CALLHOME, AMI, and DIHARD II datasets.