Tatiana Serebryakova/iStock/Getty Images Plus via Getty Images
Follow ZDNET: Add us as a preferred source on Google.
ZDNET's key takeaways
Stable Audio 2.5 is designed to help brands build a "sonic identity."
The model was trained on a fully licensed dataset.
Custom tracks can be used in ads, retail locations, and elsewhere.
Stability AI just made it easier for brands to create custom, AI-generated audio, thereby negating the need to spend time and money on elaborate recording and production processes.
The UK-based company unveiled Stable Audio 2.5 on Wednesday, describing the new model on their website as "the first audio generation model designed specifically for enterprise-grade sound-production."
Also: 4 ways machines will automate your business - and it's no hype, says Gartner
Stable Audio 2.5 is intended to help brands create high-quality and fully licensed audio clips that can be used across a variety of channels to strengthen their "sonic identity" -- that is, the collection of sounds associated with their unique marketing and branding.
"To help enterprises create the right sound, our team can fine-tune Stable Audio models on an organization's sound library, embedding signature brand audio into custom generative workflows," Stability writes. "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 Stable Audio 2.5 do?
Stability AI said its new model can create custom musical tracks of up to three minutes within seconds. It can also go beyond monotone jingles to create "multipart compositions," complete with an intro, a middle section, and an outro.
Audio 2.5 can also respond to natural language prompt specifications, like "uplifting," which modify the tone and tenor of its output (similarly 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 is best - hear my results
There's also an "inpainting" feature, enabling users to upload a snippet of their own audio, which the model will then automatically build upon. Stability AI's content moderation system will, however, reject any copyrighted material that gets 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's NotebookLM now lets you customize your AI podcasts in tone and length
That's important to note given the company is currently being sued by a group of artists who claim that it illegally used copyrighted materials in order to train Stable Diffusion, its flagship image-generating model, which was released in 2022. (Other AI companies, including Midjourney, are also targeted in the lawsuit.)
Try it for yourself
You can try Stable Audio 2.5 here. There's a free option that comes with 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.