Stability AI has added another audio product to its lineup, releasing the open-source text-to-audio generator Stable Audio Open 1.0 for sound design. The new model can generate up to 47 seconds of samples and sound effects, including drum beats, instrument riffs, ambient sounds, foley and production elements. It also allows for adapting variations and changing the style of audio samples. Stability AI — best known for the image generator Stable Diffusion — in September released Stable Audio, a commercial product that can generate sophisticated music tracks of up to three minutes.

The UK startup, founded in 2019, has been in a state of investor revolt that resulted in the March resignation of CEO Emad Mostaque and a relocation from its Notting Hill headquarters, but continues taking care of business by releasing new products. This latest joins “the company’s broader portfolio, which includes multiple models for code, text and audio,” writes VentureBeat.

Stable Audio Open has a narrower focus than Stable Audio, but offers some useful features. Users can fine-tune the model with their own custom audio data. “For example, a drummer could fine-tune on samples of their own drum recordings to generate new beats,” the company says in a news announcement.

Although the company calls it an “open” model, VentureBeat notes that it is “not technically open source,” since “rather than using an actual Open Source Initiative (OSI) approved license, Stable Audio Open is available to users under the Stability AI non-commercial research community agreement license,” which provides liberal access to the model, but restricts usage.

“The fine-tuning of Stable Audio is enabled via the Stable Audio Tools library, which is licensed under an actual open-source license,” VentureBeat says.

Unlike Stable Audio, “which is trained on copyrighted music sources, Stable Audio Open has only been trained on 486,000 samples from free music libraries such as Free Music Archive and Freesound, so the recordings are unlikely to be of the same quality,” SiliconANGLE reports. Stability AI says Stable Audio Open’s training ensures no copyrights have been infringed, allowing it to be offered free.

Stability AI has made the Stable Audio Open Model weights available on Hugging Face.

“As music generators including Stability’s gain in popularity, copyright — and the ways in which some creators of generators might be abusing it — is becoming a central point of focus,” explains TechCrunch, noting that in March, Tennessee became the first state to enact a law protecting copyrighted music against generative AI.