Stable Audio Open

Stable Audio Open enables anyone to generate high-quality audio clips of up to 47 seconds from simple text prompts. Its specialized training makes it ideal for creating drum beats, instrumental riffs, ambient sounds, Foley recordings, and other audio samples for music production and sound design. A key benefit of this open-source release is that users can fine-tune the model on their own custom audio data. For example, a drummer could fine-tune it on samples of their own drum recordings to generate new rhythms. **How is it different from Stable Audio?** Stable Audio produces high-quality full tracks with coherent musical structures up to three minutes long, along with advanced features like text-to-audio generation and cohesive multi-part musical compositions. On the other hand, Stable Audio Open specializes in audio samples, sound effects, and production elements. While it can generate short musical clips, it is not optimized for full songs, melodies, or vocals. This open model offers a glimpse into generative AI for sound design while prioritizing responsible development alongside the creative community. The new model was trained on audio data from Freesound and the Free Music Archive. This allows us to create an open audio model while respecting creators' rights.

Stable Audio Open is an open-source text-to-audio model that generates high-quality audio samples up to 47 seconds long from text prompts, primarily for music production and sound design.