Deep neural network model advancements have led to notable improvements in false sound creation. Therefore, it becomes more crucial than ever to create foolproof yet lightweight systems for false sound detection. Training with graphics, videos, and audio has gotten easier and more user-friendly. We still must deal with some dangers and drawbacks. In this post, we'll talk about the frightening sound produced by deep fakes, a term that's very common in cutting-edge technology. The use of false sound can be detrimental and have an impact on human existence either directly or indirectly. For instance, deep learning navigation is used by Google Maps; if changed, we will be redirected. There were numerous papers on how to differentiate real or false sound. To accomplish the task, Python and deep learning were employed. The input for this work is audio or video data, and the model was trained for specifically recognizable features for voice production and voice identification. The accuracy between real and fake is determined using the deep learning technique.