Amplifying Artifacts with Speech Enhancement in Voice Anti-spoofing

TL;DR

This paper presents a novel pipeline that amplifies artifacts in spoofed speech using noise addition and speech enhancement, significantly improving the accuracy of voice anti-spoofing systems.

Contribution

We introduce a model-agnostic artifact amplification pipeline that enhances spoof detection by leveraging speech enhancement and noise manipulation techniques.

Findings

Improves spoof detection accuracy by up to 44.44% on ASVspoof2019.

Enhances detection performance by 26.34% on ASVspoof2021.

Compatible with various speech enhancement models and countermeasure architectures.

Abstract

Spoofed utterances always contain artifacts introduced by generative models. While several countermeasures have been proposed to detect spoofed utterances, most primarily focus on architectural improvements. In this work, we investigate how artifacts remain hidden in spoofed speech and how to enhance their presence. We propose a model-agnostic pipeline that amplifies artifacts using speech enhancement and various types of noise. Our approach consists of three key steps: noise addition, noise extraction, and noise amplification. First, we introduce noise into the raw speech. Then, we apply speech enhancement to extract the entangled noise and artifacts. Finally, we amplify these extracted features. Moreover, our pipeline is compatible with different speech enhancement models and countermeasure architectures. Our method improves spoof detection performance by up to 44.44\% on ASVspoof2019 and 26.34\% on ASVspoof2021.