FFT and Machine Learning Application on Major Chord Recognition

Abstract

The purpose of this project was to use theory of Fast Fourier Transforms and a machine learning algorithm called Treebagger in order to process and recognize musical chords. Four musical chords were selected: C, D, G, and A. To be successful, the program should recognize chords in various octaves and musical inversions. Our goal is to use digital signal processing techniques such as sampling and frequency decomposition to preprocess audio files for input and train into the machine learning algorithms. For our application, we used MATLAB's machine learning tool Treebagger.