Abstract

Musical cryptography is a technique in which plain text messages are enciphered into a musical composition. Recently, a surge of music composition by means of machine learning have produced natural-sounding music that can be deemed as composed by humans. The combination of machine-generated music and enciphering a message into the composition is a logical step in musical cryptography. Outlined in this thesis is a method that incorporates the use of a specific type of recurrent neural network, Long Short-Term Memory, and a variant of the substitution cipher to form of symmetric-key cryptography system. Exploration was also completed to determine how the architecture of the network and the dataset used can modify the output of the encryption.

Thesis Completion

2020

Semester

Spring

Thesis Chair/Advisor

Leinecker, Richard

Degree

Bachelor of Science (B.S.)

College

College of Engineering and Computer Science

Department

Computer Science

Language

English

Access Status

Open Access

Release Date

5-1-2020

Share

COinS