A text-to-music model capable of generating high-quality audio samples using a new Non-AutoRegressive transformation method
