I was hoping that this would be encoding the data in the audible, data-sounding part of the sound.

The technique described could be used on any bass-heavy music, and is in no way related to dubstep or its data-sounding-ness.