DiffSinger
DiffSinger is an open-source AI based singing vocal synthesizer software which uses uses diffusion probabilistic models to create realistic singing voices. It is typically integrated into forks of OpenUTAU.
View wiki