Stable Audio offers three pricing tiers, including a free version, a professional level at $11.99, and an enterprise subscription for customised usage plans. The platform has been trained using over 800,000 audio files, covering 19,500 hours of diverse sounds. Despite the competition, Stability AI's offering stands out for its flexibility in song length and its extensive training dataset.
Key takeaways:
- Stability AI, a London-based company, has launched its first text-to-audio platform, Stable Audio, which allows users to generate personalized audio tracks. The platform can produce songs of up to 90 seconds in length.
- The Stable Audio platform uses a diffusion model, similar to the one used in the company's image platform, Stable Diffusion. However, it has been trained with audio data instead of images, allowing users to generate songs or background audio of any length.
- Stable Audio has been trained using an extensive dataset of over 800,000 audio files, including music, sound effects, and individual instrument stems. The dataset also includes text metadata from AudioSparx, a stock music licensing company.
- Stability Audio offers three pricing tiers for users: a free version, a Professional level priced at $11.99, and an Enterprise subscription for companies seeking customized usage plans and pricing structures.