Professor Roger B. Dannenberg Co-creator of Audacity! Future in AI and Music Humans are driven by an innate desire to achieve automation, this includes the field of music Three main areas of exploration create sound Generate new type of sound by incorporating computer to make new instruments Example Utilize robotics to preform instruments Create electronic instruments such as electric piano and guitar create music Computer-generated music score re-create and extend music performance practice By inputting a huge database of repertoires by a certain composer and using a machine learning model to have the computer “learn” the style of the composer and to create music in the style of deceased artists Combination of performance model and instrument model to create performances from music scores.

In the fall of 2021, I joined Aspiring Scholars Directed Research Program (ASDRP) under its Computer Science & Engineering department and more specifically, data science, with my mentor Suresh Subramaniam. Over the course of three month, my group and I put together a natural language project that can predict the next word given an initial string and presented our findings in the program’s weekly Colloquia meeting to an audience of around 200 people.

Why Music Dataset? Music Dataset is an free, public online platform that hopes to solve the subjectivity in music by collecting responses of abstract components in classical music from a variety of audiences. It seeks to take advantage of the online setting where people from all over the world can be easily connected through one project. The viewers are asked to place descriptive tags onto music according to the unique response that the particular repertoire evokes in them.

As a rising high school junior, I was honored to be part of a highly selective summer research program: Summer Research Academies at UC Santa Barbara in the summer of 2021. The program has an overall acceptance rate of 18.9% and I was one of five girls in a group of 26 students enrolled in its 9th track called Machine Intelligence: An Introduction to Optimization and Machine learning. During the four-week program, my two teammates and I were challenged to create a machine learning project.

My favorite hobbies revolve around music as I enjoy playing the piano, the flute, and dancing. Additionally, I enjoy building both tangible and virtual projects. In the future, I would like to major in Computer Science, and currently am most appealed in the field of machine learning. Thus, I would like to combine my interests to create a machine learning project that would benefit music education. Below are some of my project brainstorm ideas as well as how they can fulfill the needs for people from a variety of backgrounds.

If you look at my other post (SRA Experience Recap 7/26), the current music generation models can be improved to become more customizable if a dataset with specific labels is created. Then, when a user prompts for a certain characteristic as input, only songs with corresponding tags would be fed into the training model, and the output result would follow the desire of the user. Possible Labels to the MusicNet datasets Composer Musical period (eg.