I have been told that decimation slows down an audio signal without changing the pitch. To slow it down by 100%, you should map each sample to twice as many samples, and then output that mapped-onto signal at the original sampling rate. But I don’t know what “map onto” means, or why this works.
I’m first hoping for a mostly verbal explanation. After that, I would like an algorithmic explanation of how decimation works for audio signal processing.