Stability AI launches text-to-music generator trained on licensed content via a partnership with music library

Stability AI, an artificial intelligence company known for its text-to-image generator, has launched Stable Audio, a text-to-music generator. The tool works by converting verbal cues from a user into an audio track. The free version allows users to create 20-second tracks, while the Pro version allows for tracks of up to 90 seconds for commercial products with fewer than 100,000 monthly active users. Commercial products with more than 100,000 users require an enterprise license. The AI algorithm behind Stable Audio was trained on licensed content through a partnership with music library AudioSparx.

The technology behind Stable Audio is similar to the one used in the Stable Diffusion image generator, using a "latent diffusion" AI architecture. The company claims to be the first in the industry to use this technique for generating audio. The company's CEO, Emad Mostaque, expressed hope that Stable Audio will empower music enthusiasts and creative professionals to generate new content with the help of AI. The company, which is valued at $1 billion, is now seeking to raise funds at a valuation of around $4 billion.

Key takeaways

Stability AI has launched Stable Audio, an AI generator that creates music from verbal cues input by a user.
The AI algorithm behind Stable Audio was trained on licensed content, in partnership with music library AudioSparx.
Stability AI, which became an AI unicorn last year, is now seeking to raise funds at a valuation of around $4 billion.
Ed Newton-Rex, founder of AI music-making platform Jukedeck and former Product Director in TikTok’s in-house AI lab, is now VP of audio at Stability AI.

Stability AI launches text-to-music generator trained on licensed content via a partnership with music library

Key takeaways

Discussion (0)