A breakbeat (break beat) refers to the sampling of particular musical movements as drum loop beats, and their subsequent use as the rhythmic basis for hip-hop and rap music. Since the 1970’s the culture of sampling and remixing has influenced and developed countless more genres, namely electronic music, while hip-hop has become the most prominent genre in the United States and been called “America’s musical export”. While the art of sampling has evolved from digging through record crates to sorting through massive bits of musical data span across the internet, the listener, DJ, and producer still must sift through hours and hours of music at times to find a suitable break sample. This thesis details the creation of a classification system and subsequent utilization tool that employs a neural network for quickly and systematically sorting through immense libraries of music. This is done for the purpose of automatically identifying and extracting suitable portions of songs that fit the classification of breakbeats, so as to more effortlessly create hip-hop music. In doing so, a modified Convolutional Neural Network, dubbed the Perceptron Audio Classifier (PAC), was created for iterative break classification based on training performed on specifically analyzed spectral frequency patterns in breaks and non-breaks. These patterns yield relational data that could potentially codify some of hip-hop’s patterns and sequences the same way we codify patterns in contemporary music theory. These identified patterns are then used as the basis for a peripheral utilization tool, dubbed the Break Identification Genius (BIG), to identify the most suitable sections of given sample sources to be used as breaks. Essentially, this thesis details the creation of a neural network that performs with an accuracy of approximately 98.9% for break classification and a peripheral tool for utilizing the results of this network to facilitate automatic break finding.