Stable Audio: AI Music & Sound Generation by Stability

Stable Audio: AI Music & Sound by Stability

Stable Audio is Stability AI's audio model — the company behind Stable Diffusion turned its diffusion expertise to music and sound. Stable Audio 2.0 produces high-quality structured tracks up to 3 minutes at 44.1kHz stereo, plus sample transformation, sound effects, and audio editing from text prompts.

The Stable Audio Open release made the base model available under a permissive license, mirroring Stability's open-source approach to image generation.

Key Features

Full-length tracks up to 3 minutes at 44.1kHz stereo
Audio-to-audio transformation: bass to guitar, beatbox to lo-fi beat
Sound effect generation from text prompts
Stable Audio Open — base model under permissive license for self-hosting
Stable Radio — 24/7 live stream of AI-generated tracks on YouTube

Ideal Use Case

Producers, sound designers, film/game audio teams, and creators who want high-quality stems and effects from text prompts — particularly when self-hosting matters (open license) or when audio-to-audio sample transformation is the workflow.

Why Use Stable Audio

Most AI music tools target song generation or composition. Stable Audio's strengths are sample transformation, sound effects, and the open-license option — different needs than Suno/Udio/AIVA serve.

FAQ

Q: Can I self-host Stable Audio? A: Yes — Stable Audio Open is released under a permissive license. The hosted Stable Audio service offers higher quality and convenience.

Q: How does Stable Audio compare to Suno? A: Suno targets full song generation with vocals; Stable Audio targets instrumental tracks, samples, and sound effects.

Q: Is the audio commercially usable? A: Yes — see the licensing terms on the hosted service for commercial use details.