Neural Style Transfer with distortion handling for Audio and Image

The advent of Artificial Intelligence (AI) has made major contributions in solving the most difficult problems in various research area. It has applications in domains such as medicine, agriculture, education, tourism, finance, etc.. Even though there is a lot of work done in Artificial Intelligence research, there is still huge hope in exploring its subdomain. As AI becomes an unstoppable technology force, it raises some difficult questions about the future role of humans in an increasingly automated world. The general consensus is that AI cannot replace human creativity. As AI makes inroads into various fields, even human never thought it would, it keeps challenging us to think in new ways. This is what inspired us to take this idea as a subject matter. This led into the technique of neural style transfer. The main idea of NST is to combine the style features of one image, with the content features of another image, called the content image to handle the distortion in images and audio inoder to increase the quality. Existing methods had distortion in their outputs. So, tuning of the loss function to provide a better output is explored in this work.