DiffRhythm
Full-length song generation model using diffusion with lyrics and style conditioning.
About
DiffRhythm is a diffusion-based model for generating full-length songs (up to 4m45s). Conditions on lyrics and musical style descriptions. Produces vocals and instrumentals together. By ASLP@NPU. Requires GPU with 12+ GB VRAM. Apache 2.0 license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- Music & Audio Generation
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Advanced (4/5)
- License
- Apache-2.0
- Minimum VRAM
- 12 GB
- Added
- Apr 3, 2026
Related Tools
Fast music generation model producing full songs with lyrics in seconds.
Open-source toolkit for audio, music, and speech generation research.
Original latent diffusion model for text-to-audio generation.
State-of-the-art music source separation model by Meta for splitting tracks.
High-fidelity neural audio codec by Meta for audio compression and tokenization.
Updated music generation model with improved quality and longer generation.