# Stable Audio

> Audio AI Model by Stability AI — available on Layer

Open-source text-to-audio model for sound effects, field recordings, and instrument samples.

Stable Audio Open from Stability AI is an open-source text-to-audio model with ~1.2B parameters, capable of generating up to 47 seconds of stereo audio at 44.1 kHz. Built on a latent diffusion transformer architecture with T5-based text conditioning, it excels at drum beats, instrument riffs, ambient sounds, and foley recordings. Trained exclusively on Creative Commons-licensed data (7,300 hours), it is ideal for sound design and music production workflows where royalty-free audio is required.

## Specifications

| Property | Value |
|----------|-------|
| Provider | Stability AI |
| Category | Audio |
| Price | 0.01 / character count per character_count |

## Capabilities

- Sound Effects

---

[Try Stable Audio](https://next.app.layer.ai/me/session/new?baseModelId=STABLE_AUDIO) | [All Audio Models](https://layer.ai/models/audio) | [All Models](https://layer.ai/models)