Audio deepfakes generated by modern text-to-speech and voice conversion systems pose serious threats to security, privacy, and trust in digital communication. This study proposes a multi-domain feature fusion framework for robust deepfake audio detection under realistic, in-the-wild conditions. A large-scale dataset comprising 31,780 audio samples, evenly split between genuine and synthetic speech and covering diverse speakers, languages, and recording environments, is utilized. Acoustic, compression-related, emotional, phase-based, prosodic, and statistical–spectral features are extracted and fused, and classification is performed using a lightweight fully connected neural network evaluated via stratified five-fold cross-validation. The proposed system achieves an average validation accuracy of 98.03% and an AUC of 0.998, demonstrating strong and stable discriminative performance. Ablation experiments and SHAP-based analysis highlight the critical role of compression-related features in enhancing robustness when combined with other feature domains. While the study focuses on audio-only detection and does not address adversarial or multimodal scenarios, the results indicate that multi-domain feature fusion offers a practical and generalizable solution for real-world deepfake audio detection, particularly in environments involving diverse codecs and synthesis techniques.
Akshat Chhatriwala, Ishita Akolkar, Namrata Shroff et al.· ITEGAM- Journal of Engineeri...· 0 citations
Traffic surveillance systems play a crucial role in intelligent transportation by enabling automated monitoring, traffic prediction, and accident detection. However, recognizing complex traffic scenarios from real-world videos remains challenging due to dynamic environments and temporal dependencies. This paper proposes a unified spatio-temporal framework that integrates YOLOv8 based object detection, convolutional neural networks for spatial feature ex traction, and long short-term memory networks for temporal modeling. Traffic videos are preprocessed to enhance visual consistency, and detected objects are transformed into structured spatial representations. Temporal dependencies across video sequences are learned using LSTM networks, and the extracted features are evaluated using multiple machine learning classifiers under different preprocessing strategies. Experimental results demonstrate that Z-score standardization improves classification performance, with Support Vector Ma chine achieving 63.27% accuracy and an F1-score of 55.66% in an eight-class traffic scenario classification task, indicating the feasibility and robustness of the proposed framework in real-world traffic environments.
Dhartee Patel, Jinal Ahir, Namrata Shroff et al.· ITEGAM- Journal of Engineeri...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
