Deepfake Content Detection using Deep Learning Techniques
Haripriya Devarajan, Issac Sathwik Vijay Palli, Azam Ahammed, Hemanth Reddy Kolagatla · 2025
The rapid advancement of deep learning has led to the widespread use of deepfake technology, enabling realistic media manipulations that pose significant ethical and security challenges. This paper presents a novel deepfake detection framework leveraging the usage of deep learning techniques i.e, Vision Transformers (ViTs) for Image classification, Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for Video Classification, Librosa and Mel-Frequency Cepstral Coefficients (MFCCs) for audio classification with enhanced computational efficiency and detection accuracy. The proposed models possessed 98 percent accuracy in detecting the fake content. By optimizing deep neural network models with hardware acceleration (leveraging the use of GPU), the system effectively distinguishes between authentic and manipulated media. This research contributes to the development of robust AI-driven security measures to counter misinformation and safeguard digital integrity.