Filtering by
- All Subjects: Jazz
- Creators: Kocour, Michael
- Creators: Electrical Engineering Program
- Member of: Barrett, The Honors College Thesis/Creative Project Collection
The purpose of this project is to create a useful tool for musicians that utilizes the harmonic content of their playing to recommend new, relevant chords to play. This is done by training various Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs) on the lead sheets of 100 different jazz standards. A total of 200 unique datasets were produced and tested, resulting in the prediction of nearly 51 million chords. A note-prediction accuracy of 82.1% and a chord-prediction accuracy of 34.5% were achieved across all datasets. Methods of data representation that were rooted in valid music theory frameworks were found to increase the efficacy of harmonic prediction by up to 6%. Optimal LSTM input sizes were also determined for each method of data representation.
For the second part of my thesis I talk about how I came to create the creative project aspect. I discuss how and why I designed the narrative that I did, and also analyzed the music I have created to illustrate how I implemented the various methods of musical storytelling that I detail in the first part of the paper. Lastly, I discuss my plans for publication and release of the creative project.
The third part of this thesis is a sample of the creative project. There is a version of the narrative that goes along with the creative project, as well as one of the eight pieces of original music on the creative project, entitled Journey.
Overall, I found that music does have meaning, it is just meaning that society ascribes to it based off of artistic intent and context, and as to whether music should mean anything, I believe that this is a question best left to be answered on an individual basis. Music can be whatever it wants to be.