Date of Award

5-2023

Document Type

Thesis

Department

Computer Science

Committee Chair/Advisor

Dr. Bart Knijnenburg

Second Advisor

Dr. Carlos Toxtli-Hernandez

Abstract

Extra-curricular learning is on the rise, and many are interested in expanding their current knowledge by utilizing the recent increase in educational technology. While many forms of educational technology exist, there are few interactive and engaging platforms that teach music theory. Apps such as Perfect Ear and MyMusicTheory are great for becoming familiar with reading music and recognizing pitches, however, they often become dry with repetition and repeated tasks. By combining existing technologies that can complete real time conversions from raw audio to MIDI, our goal was to gather information such as harmonies, key and compatible chords from the user’s input. Using this data we aimed to create dynamic lesson plans based on user input, rather than using the same repetitive prompts from overused question pools. We were successfully able to generate these lesson plans, however, the lesson plans that we were able to create are somewhat limited. Given time restraints, we struggled to implement the pruning of audio input to match the desired lesson plans, as the recorded notes must match the correct format to generate a successful plan. Furthermore, we were not able to train a reliable voice model to recognize notes before the project was due. Though not fully complete, we successfully created a prototype dynamic lesson plan that can potentially engage users and assist in the learning of music theory if implemented in future technology.

Download

Share

COinS