A DEEP LEARNING FRAMEWORK FOR REAL TIME DEEPFAKE DETECTION ACROSS SOCIAL MEDIA PLATFORMS

With the rapid advancement of artificial intelligence, deepfake technology has emerged as a significant challenge to the authenticity and reliability of digital media. In this paper, a multimodal deepfake detection system has been proposed to effectively identify manipulated content across image, video, and audio domains. The proposed approach integrates deep learning techniques such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and audio-based feature extraction methods to capture spatial, temporal, and frequency-level characteristics.          The experimental analysis highlights that combining multiple data modalities significantly improves detection performance compared to traditional single-modality approaches. The use of a fusion mechanism enables the system to detect inconsistencies between visual and audio signals, thereby enhancing classification accuracy and robustness. The proposed framework demonstrates its potential for practical applications in areas such as social media monitoring, digital forensics, and content verification.        In future work, the system can be further enhanced by incorporating larger and more diverse datasets, improving real-time processing capabilities, and exploring advanced deep learning architectures. These improvements can contribute to developing more efficient and scalable solutions for combating deepfake-related threats in real-world scenarios.

[1] P. Zhou, X. Han, V. I. Morariu and L. S. Davis (2017), “Two-Stream Neural Networks for Tampered Face Detection,” Proceedings Of The IEEE Conference On Computer Vision Pattern Recognition Workshops (CVPRW), Pp. 1831–1839. [2] K. O. Jha and R. Singh (2023), “A Comprehensive Review on Deepfake Detection Using Deep Learning Techniques,”  International Journal of Advanced Research in Computer Science (IJARCS), Vol. 13, No. 4, pp. 50–57. [3] Y. Li, X. Yang, P. Sun, H. Qi and S. Lyu (2020), “Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics,”  Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), pp. 3204–3213. [4] Hamza, A. R. R. Javed, F. Iqbal, N. Kryvinska, A. S. Almadhor, Z. Jalil and R. Borghol (2022), “ Deepfake Audio Detection via MFCC Features Using Machine Learning,” IEEE Access, Vol. 10, pp. 134018–134028. [5] H. Wadhwa, M. Rubinstein, F. Durand and W. T. Freeman (2013), “ Eulerian Video Magnification for Revealing Subtle Changes in the World,” ACM Transactions on Graphics , Vol. 32, No. 4, pp. 1–8. [6] S. K. Gupta and N. Patel (2024), “Multimodal Fake Media Detection Using Hybrid Deep Neural Networks,”  International Research Journal of Modernization in Engineering , Technology and Science (IJRMETS) , Vol. 5, No. 8, pp. 101–108.