Stability AI releases Stability Audio 3.0 for generating longer songs

In short: Stability AI has launched Stability Audio 3.0, a group of AI music models that can create songs up to 6 minutes and 20 seconds long.

What happened

Stability AI, the company known for the image tool Stable Diffusion, announced a new set of audio models called Stability Audio 3.0. An AI model is software that learns patterns from lots of examples, like a student learning by listening to many songs.

Stability Audio 3.0 includes four versions: small SFX, small, medium, and large. The two small models are aimed at generating sound effects and music directly on a device, and they can make tracks up to two minutes long.

The medium and large models can generate full musical pieces up to 6 minutes and 20 seconds, and the company says they can keep a consistent structure and melody over that longer time. Stability AI said this is more than double the length supported by its earlier Stability Audio 2.0 model.

Stability AI is releasing the small SFX, small, and medium versions with “open weights,” meaning the core files are available so others can use and modify them (like getting the recipe, not just the meal). The large version is only available through paid options, and businesses with more than $1 million in revenue need an enterprise license.

Why it matters

Tools that can generate longer music could be useful for creators who need background tracks, rough drafts, or quick ideas. At the same time, the music industry is watching questions about training data and licensing closely, and Stability AI says these models are built on fully licensed data.

Source: TechCrunch AI