Separation Guided Speaker Diarization in Realistic Mismatched Conditions

TL;DR

This paper introduces a separation guided speaker diarization approach that combines speech separation and clustering to better handle overlapping speech in realistic conditions, significantly reducing diarization errors.

Contribution

It proposes a novel SGSD framework that effectively integrates speech separation with clustering, addressing the limitations of conventional methods in overlapping speech scenarios under mismatched conditions.

Findings

SGSD significantly reduces diarization error rates by over 20%.

Separation-based processing improves handling of overlapping speech in realistic data.

The approach outperforms state-of-the-art clustering-based diarization systems.

Abstract

We propose a separation guided speaker diarization (SGSD) approach by fully utilizing a complementarity of speech separation and speaker clustering. Since the conventional clustering-based speaker diarization (CSD) approach cannot well handle overlapping speech segments, we investigate, in this study, separation-based speaker diarization (SSD) which inherently has the potential to handle the speaker overlap regions. Our preliminary analysis shows that the state-of-the-art Conv-TasNet based speech separation, which works quite well on the simulation data, is unstable in realistic conversational speech due to the high mismatch speaking styles in simulated training speech and read speech. In doing so, separation-based processing can assist CSD in handling the overlapping speech segments under the realistic mismatched conditions. Specifically, several strategies are designed to select between the results of SSD and CSD systems based on an analysis of the instability of the SSD system performances. Experiments on the conversational telephone speech (CTS) data from DIHARD-III Challenge show that the proposed SGSD system can significantly improve the performance of state-of-the-art CSD systems, yielding relative diarization error rate reductions of 20.2% and 20.8% on the development set and evaluation set, respectively.