Instrument identification with convolutional neural networks

Kevin Rutter
kr7700@ship.edu
Faculty sponsor: Dr. Dudley Girard
College of Arts and Sciences
Engineering

This project identifies the musical instrument in an audio snippet, using short-time Fourier transform and a convolutional neural network. The accuracy of the system was measured using different filter sizes in the network. Increasing the filter size resulted in a statistically significant decrease in accuracy.