Lightweight Resolution-Aware Audio Deepfake Detection via Cross-Scale Attention and Consistency Learning

TL;DR

This paper introduces a lightweight, resolution-aware audio deepfake detection method that uses cross-scale attention and consistency learning to improve robustness across various challenging conditions and datasets.

Contribution

It presents a novel multi-resolution spectral modeling framework with cross-scale attention and consistency learning, outperforming existing methods in robustness and efficiency.

Findings

Achieves near-perfect detection on ASVspoof LA (EER 0.16%)

Maintains high robustness across multiple datasets and conditions

Requires only 159k parameters and less than 1 GFLOP per inference

Abstract

Audio deepfake detection has become increasingly challenging due to rapid advances in speech synthesis and voice conversion technologies, particularly under channel distortions, replay attacks, and real-world recording conditions. This paper proposes a resolution-aware audio deepfake detection framework that explicitly models and aligns multi-resolution spectral representations through cross-scale attention and consistency learning. Unlike conventional single-resolution or implicit feature-fusion approaches, the proposed method enforces agreement across complementary time--frequency scales. The proposed framework is evaluated on three representative benchmarks: ASVspoof 2019 (LA and PA), the Fake-or-Real (FoR) dataset, and the In-the-Wild Audio Deepfake dataset under a speaker-disjoint protocol. The method achieves near-perfect performance on ASVspoof LA (EER 0.16%), strong robustness on ASVspoof PA (EER 5.09%), FoR rerecorded audio (EER 4.54%), and in-the-wild deepfakes (AUC 0.98, EER 4.81%), significantly outperforming single-resolution and non-attention baselines under challenging conditions. The proposed model remains lightweight and efficient, requiring only 159k parameters and less than 1~GFLOP per inference, making it suitable for practical deployment. Comprehensive ablation studies confirm the critical contributions of cross-scale attention and consistency learning, while gradient-based interpretability analysis reveals that the model learns resolution-consistent and semantically meaningful spectral cues across diverse spoofing conditions. These results demonstrate that explicit cross-resolution modeling provides a principled, robust, and scalable foundation for next-generation audio deepfake detection systems.