Stable Audio 2.0 is a new AI model capable of generating high-quality full-length tracks with coherent musical structure, lasting up to three minutes at a stereo frequency of 44.1 kHz, all from a single description. The tool can generate intros, full chord progressions, and the final part of the track. Unlike its previous version, which was limited to only 90 seconds of music generation, this new version is free and publicly available.

Working with the tool is done through text queries, but there is an option to upload your own audio, and after analyzing the provided track, the system will create something similar. It's noted that this is still AI-generated music, so the result doesn't reach human-level quality, but the progress is quite noticeable. One of the noted issues is the addition of strange vocals to the tracks, and certain errors are allowed, which are considered normal at this stage of development.