ASLP-lab/DiffRhythm
A latent diffusion model for generating full-length songs with vocals and instrumental tracks.
Not currently ranked — collecting fresh signals.
star history
DiffRhythm is an open-source diffusion-based music generation model that creates complete songs with both vocals and instrumental tracks. It uses a latent diffusion architecture for efficient end-to-end generation of full-length songs. The project includes multiple versions (DiffRhythm, DiffRhythm+, and DiffRhythm 2) and provides Hugging Face space demos and audio samples.
Frequently asked
- What is ASLP-lab/DiffRhythm?
- A latent diffusion model for generating full-length songs with vocals and instrumental tracks.
- Is DiffRhythm open source?
- Yes — ASLP-lab/DiffRhythm is open source, released under the Apache-2.0 license.
- What language is DiffRhythm written in?
- ASLP-lab/DiffRhythm is primarily written in Python.
- How popular is DiffRhythm?
- ASLP-lab/DiffRhythm has 2.3k stars on GitHub.
- Where can I find DiffRhythm?
- ASLP-lab/DiffRhythm is on GitHub at https://github.com/ASLP-lab/DiffRhythm.