Stable Audio Open

Best for text to sound effects and short loops.

Stability AI's open text-to-audio model, released with the stable-audio-tools training and inference code. A solo maker types a description like a metal door slam with reverb, or a short 8-bit victory jingle, and gets up to 47 seconds of 44.1 kHz stereo audio, which makes it a practical sound-effect and stinger generator rather than a full-song tool. The code is MIT; the weights fall under the Stability AI Community License, which is free below $1M annual revenue.

Website ↗ Repository ↗ Docs ↗ Models ↗

0props 0 comments

Stable Audio Open — Picture screenshot of the model page on huggingface.co.

Games made with it

No Builder Notes name it yet.

Comments 0

Say what you used. No tool-shaming, no scores, no drive-bys.

Used this on a game? Say how it went. The people who make it read here.

Comments and props are for members. Joining takes a minute and costs nothing.