Unlike the commercial Stable Audio product, which produces full tracks with coherent musical structure up to three minutes in length, Stable Audio Open specializes in audio samples, sound effects, and production elements. The model was trained on audio data from FreeSound and the Free Music Archive, respecting creator rights. The model weights are available on Hugging Face, and the creators encourage feedback from users.
Key takeaways:
- Stable Audio Open is an open source text-to-audio model that can generate up to 47 seconds of audio samples and sound effects.
- The model allows users to create a variety of sounds including drum beats, instrument riffs, ambient sounds, and foley recordings.
- Unlike the commercial Stable Audio product, Stable Audio Open specializes in audio samples and sound effects, and is not optimized for full songs or vocals.
- The model weights for Stable Audio Open are available on Hugging Face, and the creators encourage feedback and exploration from sound designers, musicians, developers, and audio enthusiasts.