arXiv cs.CL· Khalid Zaman, Melike Sah, Anuwat Chaiwongyenc, Cem Direkoglu·· 6 天前AI 评分22
量子视觉理论用于音频分类：QV-CNN 与 QV-ViT 检测深度伪造语音
Quantum Vision Theory Applied to Audio Classification for Deepfake Speech Detection
AI 导读
研究者将量子视觉（QV）理论用于深度伪造语音检测，把 STFT、Mel-spectrograms 和 MFCC 经 QV block 转换为信息波，训练 QV-CNN 与 QV-ViT。
正文 · 原文
Abstract:We propose Quantum Vision (QV) theory as a new perspective for deep learning-based audio classification, applied to deepfake speech detection. Inspired by particle-wave duality in quantum physics, QV theory is based on the idea that data can be represented not only in its observable, collapsed form, but also as information waves. In conventional deep learning, models are trained directly on these collapsed representations, such as images. In QV theory, inputs are first transformed into information waves using a QV block, and then fed into deep learning models for classification. QV-based models improve performance in image classification compared to their non-QV counterparts. What if QV theory is applied speech spectrograms for audio classification tasks? This is the motivation and novelty of the proposed approach. In this work, Short-Time Fourier Transform (STFT), Mel-spectrograms, and Mel-Frequency Cepstral Coefficients (MFCC) of speech signals are converted into information waves using the proposed QV block and used to train QV-based Convolutional Neural Networks (QV-CNN) and QV-based Vision Transformers (QV-ViT). Extensive experiments are conducted on the ASVSpoof dataset for deepfake speech classification. The results show that QV-CNN and QV-ViT consistently outperform standard CNN and ViT models, achieving higher classification accuracy and improved robustness in distinguishing genuine and spoofed speech. Moreover, the QV-CNN model using MFCC features achieves the best overall performance on the ASVspoof dataset, with an accuracy of 94.20% and an EER of 9.04%, while the QV-CNN with Mel-spectrograms attains the highest accuracy of 94.57%. These findings demonstrate that QV theory is an effective and promising approach for audio deepfake detection and opens new directions for quantum-inspired learning in audio perception tasks.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2604.08104 [cs.CL] |
| (or arXiv:2604.08104v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2604.08104 arXiv-issued DOI via DataCite |
Submission history
From: Khalid Zaman [view email]
[v1]
Thu, 9 Apr 2026 11:22:40 UTC (11,046 KB)
[v2]
Fri, 9 Oct 2026 11:56:24 UTC (9,800 KB)
来源：arXiv cs.CL · arxiv.org