Audio Deepfake Detection with Self-Supervised WavLM and Multi-Fusion Attentive Classifier

TL;DR

This paper introduces a novel audio deepfake detection method combining self-supervised WavLM features with a Multi-Fusion Attentive classifier, achieving state-of-the-art results on multiple datasets.

Contribution

It is the first to utilize WavLM features for spoofing detection and proposes a Multi-Fusion Attentive classifier to enhance feature representation.

Findings

Achieves state-of-the-art results on ASVspoof 2021 DF set.

Provides competitive results on ASVspoof 2019 and 2021 LA sets.

Demonstrates the effectiveness of combining self-supervised features with multi-level attention.

Abstract

With the rapid development of speech synthesis and voice conversion technologies, Audio Deepfake has become a serious threat to the Automatic Speaker Verification (ASV) system. Numerous countermeasures are proposed to detect this type of attack. In this paper, we report our efforts to combine the self-supervised WavLM model and Multi-Fusion Attentive classifier for audio deepfake detection. Our method exploits the WavLM model to extract features that are more conducive to spoofing detection for the first time. Then, we propose a novel Multi-Fusion Attentive (MFA) classifier based on the Attentive Statistics Pooling (ASP) layer. The MFA captures the complementary information of audio features at both time and layer levels. Experiments demonstrate that our methods achieve state-of-the-art results on the ASVspoof 2021 DF set and provide competitive results on the ASVspoof 2019 and 2021 LA set.