Reliable Audio Deepfake Detection in Variable Conditions via Quantum-Kernel SVMs

2025-12-21

AI Summary

To address poor generalization and high false alarm rates in audio deepfake detection under scarce labeled data and highly variable recording conditions, this paper proposes a quantum kernel support vector machine (QK-SVM) method grounded in classically simulatable quantum feature maps. Without altering the feature extraction pipeline (e.g., Mel-spectrogram preprocessing) or introducing additional trainable parameters, our approach replaces the classical SVM kernel with a quantum-derived kernel, thereby preserving model lightweightness while substantially improving cross-domain robustness. Evaluated on four major benchmarks—ASVspoof 2019 LA, ASVspoof 5 (2024), ADD23, and In-the-Wild—QK-SVM achieves consistently lower equal error rates (EER) than classical SVM, with maximum EER reduction of 56.9% on ADD23. Absolute false alarm rate reductions range from 0.053 to 0.116. This work provides the first empirical validation that quantum kernels enhance inter-class separability in audio deepfake detection.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Adversarial Attacks & Robustness
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Detecting synthetic speech is challenging when labeled data are scarce and recording conditions vary. Existing end-to-end deep models often overfit or fail to generalize, and while kernel methods can remain competitive, their performance heavily depends on the chosen kernel. Here, we show that using a quantum kernel in audio deepfake detection reduces falsepositive rates without increasing model size. Quantum feature maps embed data into high-dimensional Hilbert spaces, enabling the use of expressive similarity measures and compact classifiers. Building on this motivation, we compare quantum-kernel SVMs (QSVMs) with classical SVMs using identical mel-spectrogram preprocessing and stratified 5-fold cross-validation across four corpora (ASVspoof 2019 LA, ASVspoof 5 (2024), ADD23, and an In-the-Wild set). QSVMs achieve consistently lower equalerror rates (EER): 0.183 vs. 0.299 on ASVspoof 5 (2024), 0.081 vs. 0.188 on ADD23, 0.346 vs. 0.399 on ASVspoof 2019, and 0.355 vs. 0.413 In-the-Wild. At the EER operating point (where FPR equals FNR), these correspond to absolute false-positiverate reductions of 0.116 (38.8%), 0.107 (56.9%), 0.053 (13.3%), and 0.058 (14.0%), respectively. We also report how consistent the results are across cross-validation folds and margin-based measures of class separation, using identical settings for both models. The only modification is the kernel; the features and SVM remain unchanged, no additional trainable parameters are introduced, and the quantum kernel is computed on a conventional computer.

Problem

• Detect synthetic speech with scarce labeled data
• Reduce false positives without increasing model size
• Generalize across variable recording conditions

Innovation

• Quantum-kernel SVMs for audio deepfake detection
• Quantum feature maps embed data into Hilbert spaces
• Reduced false-positive rates without increasing model size