I'm new here so I hope I'm posting in the right section. I'm trying to implement (using C#) a function to retrieve MFCC files from a given audio file. For the sake of ease, the file will be an uncompressed WAV file, so no encoding/decoding to worry about. However, I'm really at a loss as to where to start regarding MFCCs. I'm obtaining the FFT using the FFTW library, however I don't really understand where to go on from here. I understand the theory behind MFCCs but translating to code is escaping me. How would I go about defining an 'overlapping frame' and implementing filters?
Thanks for any help or pointers. I'm really at my wits end :S lol