Follow zdnet: Add us as a preferred source on Google.
ZDNET’s major takeovers
- Stable Audio 2.5 is designed to help brands create a “sound identity.”
- The model was trained on a fully licensed dataset.
- Custom songs can be used in commercials, retail outlets, and more.
Stability AI just made it easier for brands to create custom AI-generated audio, negating the need to spend time and money on complex recording and production processes.
The UK-based company unveiled the Stable Audio 2.5 on Wednesday, describing the new model on their website as “the first generation of audio specifically designed to produce enterprise-quality sound.”
Also: 4 ways machines will automate your business — and it’s not hype, says Gartner
Stable Audio 2.5 is designed to help brands create high-quality and fully licensed audio clips that can be used across channels to strengthen their “sonic identity” – that is, a collection of sounds associated with their unique marketing and branding.
“To help companies create the right sound, our team can fine-tune stable audio models within an organization’s sound library, incorporating signature brand audio into custom generative workflows,” writes Stability. “This ensures that the music or soundscape is uniquely recognizable as part of a brand’s sonic identity or creative guidelines for a project.”
What can a stable Audio 2.5 do?
Stability AI said its new model can create custom music tracks of up to three minutes in a matter of seconds. It can also go beyond monotonous jingles to create a “multi-part composition,” complete with an intro, middle section, and an outro.
Audio 2.5 can also respond to natural language prompt specifications, such as “boosts,” which modify the pitch and tenor of its output (similar to new features offered in text-to-speech models from companies like Elevenlabs).
Also: I tested 3 text-to-speech AI models to see which one is best—hear my results
There is also an “inspect” feature that allows users to upload a snippet of their own audio, which the model will automatically build on. However, Stability AI’s content moderation system will reject any copyrighted material that is uploaded.
“Like all Stable Audio models, Stable Audio 2.5 is commercially safe and trained on a fully licensed dataset,” Stability AI wrote on its website.
Also: Google Notebook now lets you customize AI podcast alerts in tone and length
This is important to note, given that the company is currently being sued by a group of artists who claim it illegally used copyrighted material to train Stable Diffusion, its flagship image-generating model that was released in 2022. (Other AI companies, including MidJourney, are also targeted in the lawsuit.)
Try it yourself
You can try out Stable Audio 2.5 here at Apvid. There’s a Free option that has a monthly limit of 10 custom tracks, a $12/month Pro option with a monthly limit of 250 tracks, and more expensive Studio and Max options.
