I need to dig into it more, but a common problem is having noticeable patterns or repeating transitions, especially if you are listening for hours at a time. But since this actually generated sound instead of trying to blend clips of audio together, I would imagine it does a pretty good job of avoiding that issue.