Disentangled Speaker Representation Learning via Mutual Information Minimization

TL;DR

This paper introduces a framework for disentangling speaker-relevant features from speaker-unrelated features using mutual information minimization, improving speaker verification performance especially in domain mismatch scenarios.

Contribution

The paper presents a novel three-stage disentanglement framework utilizing MI minimization with CLUB, and demonstrates its effectiveness through experiments on FFSVC2022 and VoxCeleb datasets.

Findings

Effective disentanglement of speaker-related and unrelated features.

Improved speaker verification accuracy after fine-tuning with the framework.

Validation on FFSVC2022 dataset shows performance gains.

Abstract

Domain mismatch problem caused by speaker-unrelated feature has been a major topic in speaker recognition. In this paper, we propose an explicit disentanglement framework to unravel speaker-relevant features from speaker-unrelated features via mutual information (MI) minimization. To achieve our goal of minimizing MI between speaker-related and speaker-unrelated features, we adopt a contrastive log-ratio upper bound (CLUB), which exploits the upper bound of MI. Our framework is constructed in a 3-stage structure. First, in the front-end encoder, input speech is encoded into shared initial embedding. Next, in the decoupling block, shared initial embedding is split into separate speaker-related and speaker-unrelated embeddings. Finally, disentanglement is conducted by MI minimization in the last stage. Experiments on Far-Field Speaker Verification Challenge 2022 (FFSVC2022) demonstrate that our proposed framework is effective for disentanglement. Also, to utilize domain-unknown datasets containing numerous speakers, we pre-trained the front-end encoder with VoxCeleb datasets. We then fine-tuned the speaker embedding model in the disentanglement framework with FFSVC 2022 dataset. The experimental results show that fine-tuning with a disentanglement framework on a existing pre-trained model is valid and can further improve performance.