Towards the Applications of Generative Adversarial Networks Beyond Images

Recently, uses of Generative Adversarial Networks (GANs) have seen tremendous advancements. The technique of GAN has been effectively utilized for the amalgamation of high-fidelity realistic images and augmentation of data, also refining image compressions, and much more. From imitating genuine expressions to discovering deep space and from linking the human-machine feeling division to introducing new art forms, GANs have shielded almost all arenas. GANSynth is the technology that is being explored for synthetization of sound waves. This study mainly focuses on the application of GAN in real-world problems and an overview of GANs, their variants, and potential applications in discrete domains and also discusses the benefits, shortcomings, and significant challenges to the practical application of GAN in various fields, including Audio and Video. The thought to ponder for researchers working in a similar field is the combination of applications of GAN in Images and Audio for its further diverse application in video synthesizing. The paper also discusses advancements of GAN applications in Audio processing besides Images and further suggests its applications in Video as well.