用量子核支持向量机，提升语音伪造检测在复杂环境下的可靠性。
Reliable Audio Deepfake Detection in Variable Conditions via Quantum-Kernel SVMs
- 采用量子核函数嵌入音频特征，不增加模型参数却提升判别能力。
- 在四个数据集上等错误率降低13%至57%，误报率显著下降。
- 无需改动模型结构，仅换核即可实现性能提升，适合部署于资源受限场景。
在标注数据稀缺且录音条件多变的情况下，合成语音检测面临挑战。现有端到端深度模型易过拟合或泛化能力差，而核方法虽具竞争力，其性能高度依赖核函数选择。本文表明，在音频伪造检测中使用量子核可降低误报率而不增加模型规模。量子特征映射将数据嵌入高维希尔伯特空间，支持表达性强的相似性度量与紧凑分类器。基于此，我们对比了量子核支持向量机（QSVM）与经典支持向量机，均采用相同的梅尔频谱图预处理和分层5折交叉验证，涵盖四个语料库：ASVspoof 2019 LA、ASVspoof 5 (2024)、ADD23 和 In-the-Wild 数据集。结果显示，QSVM 在所有数据集上等错误率（EER）更低：在 ASVspoof 5 (2024) 上为 0.183（经典为 0.299），在 ADD23 上为 0.081（经典为 0.188），在 ASVspoof 2019 上为 0.346（经典为 0.399），在 In-the-Wild 上为 0.355（经典为 0.413）。在等错误率点（此时假正率等于假负率），误报率绝对降低分别为 0.116（38.8%）、0.107（56.9%）、0.053（13.3%）和 0.058（14.0%）。同时报告了跨验证折间结果一致性及基于间隔的类别分离度量，两模型设置完全一致。唯一区别是核函数，特征与支持向量机保持不变，无新增可训练参数，量子核在传统计算机上计算。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。