The main use case for 1.5x, 2x speed would be for listening to podcasts or other learning material.
Besides pre-encoding these files, the variable playback would need to be processed on device which would take up processing power.
Has anyone heard of future support for this?
