Junkai Wu

PhD Student at the University of Washington

Junkai Wu

Hi! My name is Junkai Wu. I’m a third year Ph.D. student in the Department of Electrical & Computer Engineering at the University of Washington, advised by Prof. Mari Ostendorf. My research is on audio-centric multimodal AI, and especially on text and multimodal LLMs that understand and generate speech, music, and sound, either end to end or as components in larger systems.

Before coming to UW, I got my B.S. in Computer Engineering from the University of Illinois Urbana-Champaign, where I worked on audio processing with Prof. Paris Smaragdis and speech processing with Prof. Mark Hasegawa-Johnson.

news

  • AVMeme Exam was accepted to COLM 2026!
  • Started my research internship with the Music AI Group at Adobe.
  • Bridging Ears and Eyes won the Best Paper Award🥇 at WASPAA 2025!
  • Website lauched :cowboy_hat_face: !

selected publications

  1. WASPAABest Paper Award
    Bridging Ears and Eyes: Analyzing Audio and Visual Large Language Models to Humans in Visible Sound Recognition and Reducing Their Sensory Gap via Cross-Modal Distillation
    Xilin Jiang*, Junkai Wu*, Vishal Choudhari, and Nima Mesgarani
    In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2025
  2. SLT
    Just ASR + LLM? A Study on Speech Large Language Models’ Ability to Identify and Understand Speaker in Spoken Dialogue
    Junkai Wu*, Xulin Fan*, Bo-Ru Lu, Xilin Jiang, and 3 more authors
    In IEEE Spoken Language Technology Workshop (SLT), 2024
  3. IWAENC
    Meta-Learning for Adaptive Filters with higher-order Frequency Dependencies
    Junkai Wu, Jonah Casebeer, Nicholas J. Bryan, and Paris Smaragdis
    In International Workshop on Acoustic Signal Enhancement (IWAENC), 2022

experience

education

teaching

  • EE P 598 · Introduction to Digital Audio
    University of Washington·Seattle, WA
    TA for the course, covering audio synthesis and sound programming in SuperCollider.