Stable Audio

Stable Audio 2.0 is an AI model for generating music using text prompts and audio inputs. Its key capability is producing full compositions up to three minutes long in 44.1 kHz stereo.

What it does

Stable Audio 2.0, developed by Stability AI, creates tracks in different genres and lengths based on your description or a musical example.

• Generate tracks up to 3 minutes
• Upload audio for processing and continuation
• Create sound effects and apply audio style transfer
• Trained on licensed data with copyright compliance (AudioSparx)
• Multi-level architecture designed to preserve musical structure

How to use it

You can describe the music you want (for example, a melodic, melancholic 80s-style ballad) and the model will generate a matching composition. You can also upload a short audio clip and have the model continue it in a similar style.

• Go to the official website
• Sign up or log in
• Enter a text prompt or upload an audio file
• Adjust generation settings
• Download the resulting track to your smartphone

Notable features

• Clear musical structure in longer generations
• Audio editing options like tempo changes and instrument adjustments
• Sound effects generation (nature, instruments, city ambience)
• Free access is available