AI-based Virtual Musical Accompaniment

The overarching goal of this project is to democratize musical accompaniment. In many musical traditions like classical piano, very few musicians get to experience the thrill of playing with a live orchestra – it is reserved for the 0.01% of pianists who win a piano competition. Our goal is to develop the technology to allow anyone, anywhere to play music with a virtual accompaniment system. This summer project will focus primarily on piano concertos, and may extend to other musical traditions based on students’ backgrounds and musical expertise. This project is a collaboration with the International Audio Labs Erlangen, a leading music and audio research lab in Germany. The internship will include an international trip to Germany to participate in a joint workshop with the Audio Labs group.

For detailed instructions on how to apply, please see our lab website. (Please apply through our lab website, not through the uro website.) Applications are due Feb 18.

Name of research group, project, or lab

Music Information Retrieval (MIR) Research Lab

Website

https://sites.google.com/g.hmc.edu/hmc-mir

Why join this research group or lab?

The HMC MIR research lab explores the intersection of music, machine learning, and signal processing. Our goal is to train machine learning and signal processing ninjas using music as a playground. Many of our lab members go on to jobs in industry or Ph.D. programs in machine learning, computer vision, and audio processing. You can read about some of our lab members here.

Representative publication

ImprovedComposerID.pdf

Logistics Information:

Artificial Intelligence

Machine Learning

Signal Processing

Student ranks applicable

First-year

Sophomore

Junior

Student qualifications

Ideal candidates will have strong programming skills, experience and demonstrated interest in machine learning and/or signal processing, and be enthusiastic about our lab’s research. Having a strong background in music is a definite plus, but not required. Candidates with expertise in music genres that our lab has not explored (e.g. hip hop dance, jazz, gospel) are especially encouraged to apply.

Fall - Part Time

Spring - Part Time

Summer - Full Time

Compensation

Academic Credit

Paid Research

Techniques learned

Students will apply modern machine learning and signal processing techniques to solve problems in music information retrieval. A sampling of our current and past projects can be found here.

Project start

Spring or Summer 2024

Contact Information:

Mentor

TJ Tsai

ttsai@hmc.edu

Principal Investigator

Name of project director or principal investigator

TJ Tsai

Email address of project director or principal investigator

ttsai@g.hmc.edu

2 sp. | 6 appl.

Fall - Part Time (+2)

Signal Processing (+2)