Multi-objective Progressive Clustering for Semi-supervised Domain Adaptation in Speaker Verification

TL;DR

This paper introduces MoPC, a novel pseudo-labeling approach for semi-supervised domain adaptation in speaker verification, utilizing multi-objective clustering and label refinement to improve accuracy on challenging datasets.

Contribution

The paper presents a new multi-objective progressive clustering method that enhances pseudo-label quality for semi-supervised domain adaptation in speaker verification.

Findings

Achieved 4.95% EER on VoxSRC 2023, ranking first.

Effective pseudo-label refinement through subcenter-purification and progressive merging.

Promising results on FFSVC dataset.

Abstract

Utilizing the pseudo-labeling algorithm with large-scale unlabeled data becomes crucial for semi-supervised domain adaptation in speaker verification tasks. In this paper, we propose a novel pseudo-labeling method named Multi-objective Progressive Clustering (MoPC), specifically designed for semi-supervised domain adaptation. Firstly, we utilize limited labeled data from the target domain to derive domain-specific descriptors based on multiple distinct objectives, namely within-graph denoising, intra-class denoising and inter-class denoising. Then, the Infomap algorithm is adopted for embedding clustering, and the descriptors are leveraged to further refine the target domain's pseudo-labels. Moreover, to further improve the quality of pseudo labels, we introduce the subcenter-purification and progressive-merging strategy for label denoising. Our proposed MoPC method achieves 4.95% EER and ranked the 1$^{st}$ place on the evaluation set of VoxSRC 2023 track 3. We also conduct additional experiments on the FFSVC dataset and yield promising results.