LAUSR

Detection of fake audio and video is a challenging problem. Deepfake is frequently used for creating fake audios and videos using deep learning techniques. Deepfakes, artificially created audiovisual interpretations can be used in many different ways; such as, damaging the repute of a celebrity, misinformation or hate speech, and it may lead to chaos in the society. Therefore, deepfake detection is of utmost important. This article presents an overview of deepfake detection models and datasets, challenges and opportunities in current methods, and provides some possible solutions. This article is mainly focused on multimodal data modalities to detect audio–visual fakes.