As long as you can get the sample data from the audio files you should be able to implement your own analysis code. I’d recommend checking out http://musicdsp.org/ for a whole bunch of dsp algorithms. You could also check out some third party libraries such as JUCE or FMOD.
I remember hearing that EPIC have hired some audio coders too, it’ll be interesting to see what they come up with.