Skip to content

Stable Audio Open 1.0

Generate stereo audio from text prompts

Visit official site

What Stable Audio Open 1.0 is

The model is described as one that generates variable-length stereo audio from text prompts for research and experimentation.

What you can do with it

  • Produce stereo music from a prompt
  • Set a start time and a total length
  • Work at a 44.1 kHz sample rate

How it works

  • The stable-audio-tools library is installed
  • Text and time conditions are defined
  • The diffusion model produces the audio

Pros & cons

✓ What we like

  • Text conditioning runs through T5 embeddings
  • The diffusion model works in the autoencoder latent space

! What to watch out for

  • The stable-audio-tools library is installed first
  • Text and time conditions are set before generation

FAQ

How long can the audio be?

The material says stereo audio up to 47 seconds is generated.

What data is cited?

The material cites Freesound and the Free Music Archive.

Last reviewed: 2026-11-15

More AI music creation tools

View all →

How we review