Deepfake is defined as a multimedia content synthetically modified or created through automatic (or barely controlled) machine learning models. The rise of deepfake technology points out the importance of accurate detection methods. In this article, we propose a deepfake detection system based on Long Short-Term Memory (LSTM) networks and the ResNext architecture. Users can upload videos for examination, which determines if they are legitimate or fake. LSTM evaluate face motions, gestures, and expressions, whereas ResNext identifies and extracts facial features and landmarks. Additionally, we provide users with an option to report suspected deepfake videos via email, facilitating community involvement in identifying fraudulent content. Moreover, our platform includes a directory of legal advocates, enabling users to seek legal support tailored to their location and needs. In conclusion, our deep learning-based deepfake video detection project represents a vital step in addressing the growing threat of digital manipulation.