[ARCHIVE] Musical instrument recognition

April 2017

Created by: Sophia Mehdizadeh, Peeravich Panlertkitsakul, and Timothy Kennedy

About the project

EECS 351 is an elective electrical engineering course on digital signal processing and analysis. For our final project, my team chose to create an instrument recognition program using MATLAB. The goal of this software was to accurately classify musical instrument samples as either flute, saxophone, or violin, although it was designed with enough generality to accommodate other instruments as well. We worked with two methods of classification, support vector machine (SVM) and convolutional neural network (CNN). We trained and implemented both of these models and quantitatively evaluated their success. This project gave me hands on experience and taught me how to extract features from signals for analysis and deep/machine learning.

Our final deliverables were a class presentation and an IEEE-style technical paper.