Unsupervised Feature Enhancement for speaker verification

TL;DR

This paper introduces an unsupervised feature enhancement method in the log-filter bank domain to improve speaker verification robustness across various adverse acoustic conditions, demonstrating significant and consistent performance gains.

Contribution

The paper presents a novel unsupervised feature enhancement technique that improves speaker verification accuracy in noisy and reverberant environments, applicable with or without data augmentation.

Findings

Significant improvements on real and simulated noisy and reverberant test sets.

Effective enhancement even without data augmentation in the verification pipeline.

Slight additional gains when combined with data augmentation during training.

Abstract

The task of making speaker verification systems robust to adverse scenarios remain a challenging and an active area of research. We developed an unsupervised feature enhancement approach in log-filter bank domain with the end goal of improving speaker verification performance. We experimented with using both real speech recorded in adverse environments and degraded speech obtained by simulation to train the enhancement systems. The effectiveness of the approach was shown by testing on several real, simulated noisy, and reverberant test sets. The approach yielded significant improvements on both real and simulated sets when data augmentation was not used in speaker verification pipeline or augmentation was used only during x-vector training. When data augmentation was used for x-vector and PLDA training, our enhancement approach yielded slight improvements.