Introduction: In the 21st century of digital technology, deepfakes are increasingly becoming a serious cause for the nation. Deepfake technology, which can generate extremely realistic fake images and movies, can be used for both creative and harmful objectives..Nowadays it is very difficult to identify which image/video/media is original or fake.

Objective: Our objective of this paper is to create a robust and reliable model that recognizes the Deepfake media using some of the advanced artificial intelligence techniques like:- Machine Learning and Deep learning classifiers

Material/methods: In our research work we have used digital tools and advanced technologies that capture real-time images, and videos as well as cameras, and microphones that are used to monitor. The data we have collected from the Kaggle repository and as well as a real-time environment for training as well as testing purposes. The Edge Devices is used for video processing and analysis purposes. Deep learning algorithms like CNN, RNN, VGG16, MTCNN, InceptionResnetV1, and the Facenet_pytroch are used to identify which one is real or fake. The extensive feature selection algorithms(Recursive Feature Elimination, PCA, CORR) are used to improve the effectiveness of the model

Results: For the effectiveness of our model, we compare the training and testing accuracy of the algorithms. The performance metrics ( Accuracy, precision, recall, and F1-score) are used for unseen environment data. Our experimental work gave an excellent result with an accuracy of 95% by MTCNN, 98% by InceptionResnetV1 98% by the Facenet_pytroch, and 92% by CNN.