Stable Diffusion is an open model of artificial intelligence that can generate images based on text. Riffusion tuned this model so that it was able to create images called spectrograms and then transform them into sound clips. Spectrograms are a kind of images that show how different frequencies sound at different time. In addition, they have created an interactive internet application, thanks to which everyone can enter their topic and the application will generate a sound clip. In addition, the application allows a smooth transition between different topics or different variants of the same topic.