The world today is in a state of unrelenting change through technology and art and computer vision is an example of other possibilities. Such an advancement is style transfer, which is a process of applying the style of an image to another. However, some issues are inherent to the conventional style transfer methods including high computational complexity, suppressed style details, and visually dissimilar content creation. To overcome such challenges, a new framework is proposed to combine Style Transfer with Diffusion process. This approach involves the use of pre-trained deep learning models in feature extraction where style and content features are merged while using diffusion process to improve the visual quality, and reduce computation time. To this end, the use of denoising processes in the framework significantly minimizes noise, and increases coherence in the synthesized images to accommodate real-time applications with high artistic resolution.