Algorithmic Composition Using Neural Networks and Keyword User Input

Madison J. Miller
Dr. Alan Garvey, Faculty Mentor

Music is able to induce emotions in humans partially based on certain musical features that can be observed. The tempo, along with many other features in a track, can change the emotions induced by that track. Affectively-driven Algorithmic Composition is a field where music is composed by a computer system to induce certain emotions in humans. Artificial Neural Networks (ANN) have been around since the 1940’s, but made a resurgence in the 1980’s due to improved processing power. They are very interesting because they are modeled after the human brain and appear to learn similarly to humans. In this project, we will be creating an ANN that will understand the muscial qualities that contribute to inducing emotion in humans. This ANN will also generate music that will be able to induce an emotion from a list of keywords.

Keywords: algorithmic composition, music generation, music perception, machine improvisation, artificial intelligence

Topic(s):Computer Science

Presentation Type: Oral Presentation

Session: TBA
Location: TBA
Time: TBA