ASLP-lab/DiffRhythm · Apply for community grant: Academic project (gpu)

DiffRhythm is an academic project that focuses on the cutting-edge field of music generation. Utilizing advanced computational techniques and algorithms, this project aims to explore the creative potential of artificial intelligence in composing and generating music. By leveraging the power of deep learning models, DiffRhythm seeks to produce a wide range of musical styles and genres, from classical to contemporary, with the goal of pushing the boundaries of what is possible in automated music creation. This project not only contributes to the academic research in music technology but also has the potential to inspire new forms of artistic expression and collaboration between humans and machines.
We hope to have a public demo page where people can visually experience the performance of DiffRhythm. Its inference requires about 25GB of GPU memory.
Another contribution of this project is the comprehensive open-sourcing of training and inference code to facilitate further development in this field. The official code repository can be found at: https://github.com/ASLP-lab/DiffRhythm.