Machine Learning Music

Jacob Mai
Jacob is an 11th grade student from Wayland High School, US. He is interested in a Machine Learning Music Generation project.

Jacob is an active member in his school’s Math, Robotics and CS clubs. He is enrolled in a Computer Science course in school and has completed several basic projects. His grades are near the top of his class with 3.8 GPA and 1370 PSAT.

Scope of the program

Overall structure

Jacob’s project is on using machine learning music generation, which is 100% his idea. In particular, he worked on predicting deleted parts of a music piece from the rest of it, e.g., predicting the Soprano part from Alto and Bass. He focused on pieces from Bach, collected data of musical notes, re-organized the data, trained several machine learning algorithms for prediction, and compared the outcomes with the original piece. He then put together a report summarizing his results in machine learning music generation and made available audio pieces where we can hear out the musical piece predicted by the Al.

Learning objectives

Data collection
Data processing
Machine learning algorithms and their comparison
Programming in Python, and presenting ML results

What challenges did he face?

Jacob had difficulties putting his dataset in the right format for training the machine learning algorithms. However, under his mentor’s guidance, he was able to solve them.

How did our program in Machine Learning Music Generation help him?

Jacob’s project focused on musical notes prediction and at the end of his program, he wrote a report summarizing his results.
He learnt a lot during the program and using machine learning algorithms, he was able to predict new music pieces that sound similar to the original pieces. One example is predicting the Soprano part from Alto and Bass.

Student’s Testimonial

“In my mentorship, I went through the process of generating music with machine learning. In two months, I became familiar with the Python programming language, was introduced to the topics of data collection, and worked with the fundamentals of machine learning.

In the end, I managed to create a model that could harmonize a melody with supporting Alto Tenor and Bass parts. Throughout the entire journey, was always very kind and supportive and was always there when I needed help. This was a very fun educational experience and I hope to use some of the skills I’ve acquired in the future.“
Jacob Mai

Matched Professor

The mentor is a Research Professor in the Department of Electrical and Computer Engineering & CyLab, Carnegie Mellon University. The mentor’s research interests are in modeling, design, and performance evaluation of engineering systems, with particular emphasis on communication systems and networks.

Recommendation letter

“I would definitely recommend Jacob without any reservation. He would do well in any top CS program based on his work ethic, intelligence, and creativity.”
Mentor from Carnegie Mellon University