Stable Audio works by generating raw audio samples for higher quality output, and was trained on over 800,000 pieces of licensed music. However, it won't allow users to generate music in the style of specific artists. The technology will be available for free, allowing 20 generations per month of up to 20 second tracks, and a $12/month Pro plan, which increases this to 500 generations and 90 second tracks.
Key takeaways:
- Stability AI has announced the public release of its Stable Audio technology, which allows users to generate short audio clips using simple text prompts.
- The technology behind Stable Audio is based on a diffusion model, similar to the one used in Stability AI's Stable Diffusion text-to-image generation AI technology.
- Stable Audio does not allow users to generate music in the style of specific artists, instead encouraging more creative approaches.
- Stable Audio will be available for free and in a $12/month Pro plan, with the free version allowing 20 generations per month of up to 20 second tracks, and the Pro version offering 500 generations and 90 second tracks.