Domain-Dependent Speaker Diarization for the Third DIHARD Challenge

TL;DR

This paper introduces a domain-dependent speaker diarization system that leverages acoustic domain identification to improve performance, achieving significant DER reductions in the DIHARD III challenge.

Contribution

The work presents a simple, efficient acoustic domain-dependent diarization approach using speaker embeddings, with i-vector based ADI outperforming x-vector methods.

Findings

i-vector based ADI outperforms x-vector in DIHARD dataset

Optimized thresholds and parameters improve diarization accuracy

Achieved 9.63% and 10.64% DER reduction in DIHARD III

Abstract

This report presents the system developed by the ABSP Laboratory team for the third DIHARD speech diarization challenge. Our main contribution in this work is to develop a simple and efficient solution for acoustic domain dependent speech diarization. We explore speaker embeddings for \emph{acoustic domain identification} (ADI) task. Our study reveals that i-vector based method achieves considerably better performance than x-vector based approach in the third DIHARD challenge dataset. Next, we integrate the ADI module with the diarization framework. The performance substantially improved over that of the baseline when we optimized the thresholds for agglomerative hierarchical clustering and the parameters for dimensionality reduction during scoring for individual acoustic domains. We achieved a relative improvement of $9.63\%$ and $10.64\%$ in DER for core and full conditions, respectively, for Track 1 of the DIHARD III evaluation set.