PadAug: Robust Speaker Verification with Simple Waveform-Level Silence Padding

TL;DR

PadAug is a simple waveform-level data augmentation technique that concatenates silence and speech segments during training, significantly improving the robustness and accuracy of speaker verification systems against silence segments.

Contribution

This paper introduces PadAug, a novel waveform-level augmentation method that enhances speaker verification robustness to silence segments, compatible with current architectures.

Findings

Achieves 5.0% relative EER reduction on VoxCeleb with ResNet34.

Systems with PadAug are robust to varying silence lengths and proportions.

Effective in improving speaker verification performance under silence segment variations.

Abstract

The presence of non-speech segments in utterances often leads to the performance degradation of speaker verification. Existing systems usually use voice activation detection as a preprocessing step to cut off long silence segments. However, short silence segments, particularly those between speech segments, still remain a problem for speaker verification. To address this issue, in this paper, we propose a simple wave-level data augmentation method, \textit{PadAug}, which aims to enhance the system's robustness to silence segments. The core idea of \textit{PadAug} is to concatenate silence segments with speech segments at the waveform level for model training. Due to its simplicity, it can be directly applied to the current state-of-the art architectures. Experimental results demonstrate the effectiveness of the proposed \textit{PadAug}. For example, applying \textit{PadAug} to ResNet34 achieves a relative equal error rate reduction of 5.0\% on the voxceleb dataset. Moreover, the \textit{PadAug} based systems are robust to different lengths and proportions of silence segments in the test data.