Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper establishes interpretable classical machine learning baselines for deepfake audio detection on the Fake-or-Real dataset, using prosodic, voice-quality, and spectral features. Statistical analysis identifies key discriminative cues like pitch variability and spectral richness. An RBF SVM achieves ~93% test accuracy and ~7% EER, demonstrating the effectiveness of classical methods and providing a strong baseline for future deep learning approaches.
You don't need deep learning to achieve 93% accuracy in detecting deepfake audio – a well-tuned SVM with interpretable acoustic features gets you there.
Deep learning has enabled highly realistic synthetic speech, raising concerns about fraud, impersonation, and disinformation. Despite rapid progress in neural detectors, transparent baselines are needed to reveal which acoustic cues reliably separate real from synthetic speech. This paper presents an interpretable classical machine learning baseline for deepfake audio detection using the Fake-or-Real (FoR) dataset. We extract prosodic, voice-quality, and spectral features from two-second clips at 44.1 kHz (high-fidelity) and 16 kHz (telephone-quality) sampling rates. Statistical analysis (ANOVA, correlation heatmaps) identifies features that differ significantly between real and fake speech. We then train multiple classifiers -- Logistic Regression, LDA, QDA, Gaussian Naive Bayes, SVMs, and GMMs -- and evaluate performance using accuracy, ROC-AUC, EER, and DET curves. Pairwise McNemar's tests confirm statistically significant differences between models. The best model, an RBF SVM, achieves ~93% test accuracy and ~7% EER on both sampling rates, while linear models reach ~75% accuracy. Feature analysis reveals that pitch variability and spectral richness (spectral centroid, bandwidth) are key discriminative cues. These results provide a strong, interpretable baseline for future deepfake audio detectors.