Teaching
Teaching Assistant
Spring 2024 ECE 411 Selected Topics in Augmented and Virtual Reality
- Gave a guest lecture on audio deepfake detection and watermarking
- Designed two new programming homework assignment on “Reverberation and Spatial Audio” and “Source Separation and Voice Cloning”
Fall 2023 ECE 477 Computer Audition
- Gave a guest lecture on python programming for audio
- Gave a guest lecture on speech technology
- Gave a guest lecture on speech anti-spoofing
- Designed one new homework assignment on speaker diarization
Spring 2023 ECE 208/408 The Art of Machine Learning
- Gave a guest lecture on SVM
- Gave a guest lecture on network training
- Co-designed four out of eight new homework assignments
Fall 2022 ECE 440 Introduction to Random Processes
Spring 2022 ECE 208/408 The Art of Machine Learning
- Gave a guest lecture on GAN
- Designed one homework question
Spring 2021 ECE 272/472 Audio Signal Processing
Fall 2020 ECE 477 Computer Audition
Spring 2020 ECE 272/472 Audio Signal Processing
- Designed one homework question
Fall 2019 ECE 216 Microprocessor & Data Conversion
Students Mentored
Kyungbok Lee, Computer Science Undergraduate@UR
Yutong (Cooper) Wen, Audio & Music Engineering Undergrad@UR, Next Move: PhD student@UIUC
Enting Zhou, Computer Science Undergraduate@UR, Next Move: Master@UCSD
Yongyi (Colin) Zang, Audio & Music Engineering Undergrad@UR, Next Move: Machine Learning Engineer@Neosensory
Siwen (Sivan) Ding, Data Science Master@Columbia, Next Move: PhD student@NYU
Abudukelimu Wuerkaixi, junior PhD student@Tsinghua
Xinhui Chen, Computer Science Master@UR, Next Move: PhD student@Lehigh
Tutorials
Personalizing Spatial Audio: Machine Learning for Personalized Head-Related Transfer Functions (HRTFs) Modeling in Gaming
2024 AES International Conference on Audio for Games. [Slides]
Others
I contributed to an ESPnet tutorial used in CMU 11751/18781 Fall 2022. ESPnet is a widely-used end-to-end speech processing toolkit. It has supported various speech-processing tasks. The goal of the tutorial is to demonstrate how to add new tasks, add new models, and create a new recipe for new tasks. We select speech anti-spoofing as the new task. [Colab]