Augmentation through Laundering Attacks for Audio Spoof Detection

TL;DR

This paper explores the effectiveness of data augmentation via laundering attacks in improving audio spoof detection systems, evaluated on diverse and challenging datasets from the ASVspoof 5 Challenge.

Contribution

It introduces a laundering attack-based data augmentation method for training audio spoof detection models and assesses its performance on the ASVspoof 5 database.

Findings

System performs poorly on certain spoofing attacks and codec conditions.

Augmentation improves robustness against some attack types.

Challenges remain in detecting specific spoofing and compression scenarios.

Abstract

Recent text-to-speech (TTS) developments have made voice cloning (VC) more realistic, affordable, and easily accessible. This has given rise to many potential abuses of this technology, including Joe Biden's New Hampshire deepfake robocall. Several methodologies have been proposed to detect such clones. However, these methodologies have been trained and evaluated on relatively clean databases. Recently, ASVspoof 5 Challenge introduced a new crowd-sourced database of diverse acoustic conditions including various spoofing attacks and codec conditions. This paper is our submission to the ASVspoof 5 Challenge and aims to investigate the performance of Audio Spoof Detection, trained using data augmentation through laundering attacks, on the ASVSpoof 5 database. The results demonstrate that our system performs worst on A18, A19, A20, A26, and A30 spoofing attacks and in the codec and compression conditions of C08, C09, and C10.