Stability AI released Stable Audio Open, an AI model for generating sounds from text descriptions trained on royalty-free recordings.
Users can fine-tune the model on their own custom audio data, but it has limitations in generating full songs, melodies, vocals, and commercial use.