Analyzing an audio file.

So what I'm doing is I'm procedurally generating a level before the user enters the playing game state. I know how to edit positions and all that fancy stuff but the main thing I'm wondering. Is it possible to analyze an audio file or waveform of an audio file, faster than its played? I know how to get the readings but how could I speed it up to were I can analyze it based off of a faster reading time?

Example: 1 minute 30 second song, the song gets split into 1 second flat readings for analyzing. How would I be able to jump from 1 second to another second without having to play it in its natural time?