Label-Efficient Self-Supervised Speaker Verification With Information Maximization and Contrastive Learning

TL;DR

This paper proposes a self-supervised learning approach for speaker verification that leverages information maximization and contrastive learning to produce robust speaker embeddings without extensive labeled data.

Contribution

It introduces a novel self-supervised framework combining information maximization and contrastive learning for speaker verification, reducing reliance on labeled datasets.

Findings

Achieves competitive results with existing methods.

Outperforms supervised baseline when fine-tuned with limited labeled data.

Effective data augmentation enhances embedding robustness.

Abstract

State-of-the-art speaker verification systems are inherently dependent on some kind of human supervision as they are trained on massive amounts of labeled data. However, manually annotating utterances is slow, expensive and not scalable to the amount of data available today. In this study, we explore self-supervised learning for speaker verification by learning representations directly from raw audio. The objective is to produce robust speaker embeddings that have small intra-speaker and large inter-speaker variance. Our approach is based on recent information maximization learning frameworks and an intensive data augmentation pre-processing step. We evaluate the ability of these methods to work without contrastive samples before showing that they achieve better performance when combined with a contrastive loss. Furthermore, we conduct experiments to show that our method reaches competitive results compared to existing techniques and can get better performances compared to a supervised baseline when fine-tuned with a small portion of labeled data.