My objective is to make a speaker recognition program using Delphi.

So far, I have managed to record my voice through a microphone into a .wav file. However, I get confused about how to process the .wav file using FFT to get the frequency domain. I have read several papers about FFT, yet they are too theoritical and when I find it difficult to put it into Delphi codes.