sure, well, as input data i mean the played note e.g.: on a guitar, received from my device over the microphone and of course as stored data i mean the saved *.wav or *.mp3 or whatever files to compare with the received one.
The device sould finally emulate an autotuner device for music instruments.
Harmonics & local acoustics would play havoc with comparison to a .wav. I suspect that you'd need to do some interesting maths to find the main underlying frequency of your note & then see if it was close enough to call "in tune".