About Me

I am Jayeon Yi, a first-year PhD Student in Computer Science at the University of Illinois Urbana-Champaign. I am deeply grateful and fortunate to be working as a research assistant for Prof. Minje Kim. My current research focus is to investigate few-shot learning of esoteric speech, audio, and music concepts. My current projects include DNN-based speech codecs – arguably an extreme (and important!) form of latent represenation learning and manipulation. I hope to be able to work on more ideas as I progress further into my degree!

Previously, I worked as an Applied Scientist Intern at Amazon, Sunnyvale. Over two summers, I had thankfully been given the chance to work on awesome projects involving speech separation or DNN-based codecs. Before this, I worked with Prof. Hun-Seok Kim on a project related to video and image compression, during my Master’s at the University of Michigan. Even before this, I was a student intern at Music and Audio Research Group, Seoul National University, under the guidance of Prof. Kyogu Lee. Projects from this period culminated in works in Speech Declipping and Music-synced game content generation, respectively presented in ICASSP 2024 and ISMIR 2023 LBD.

In my free time, I make music for fun. I have experience providing or selling some of my music to labels or games under a pseudonym.

Here’s a CV, which includes a list of publications.