AudioCraft (Meta)
AudioCraft is a PyTorch library developed to conduct deep learning research in the field of audio production. The library contains both inference and training codes for AI productive models such as MusicGen, AudioGen, EnCodec, Multi Band Diffusion, MAGNeT, AudioSeal, MusicGen Style and JASCO. It provides configurable components and training pipelines for audio and music processing processes.
facebookresearch · Music & Audio Technology
Common commercial products in this category: Adobe Audition, Ableton Live, Spotify
- Using MusicGen, MAGNeT or JASCO models, you can produce music with text-based commands, chords, and melodies.
- Using the AudioGen model, you can create a variety of sound effects and environmental sounds through text inputs.
- With EnCodec and Multi Band Diffusion components, you can perform high fidelity neural audio encoding and data processing.
- • NeutralGitHub Issue
Add generate_continuation to app.py
“It would be cool if you could add ```generate_continuation``` for the Gradio app. I was able to hack this together myself by adding a ```gr.Slider``` then modifying the ```predict``` function. I had it so you could choose where to generate from in the song (with the continuation…”
- ↓ NegativeGitHub Issue
There was a problem when trying to write in your cache folder (C:\Users\Lenovo\.cache). You should set the environment v
“I have problems in running the audiocraft. Every time i run the command, python demos\musicgen_app.py, it says that i need to fix the cache. can someone help me? The error is: There was a problem when trying to write in your cache folder (C:\Users\Lenovo\.cache). You should set…”
Verified against README · Oct 26, 2026 · source