Music AI · Machine Learning · Music Technology
Yuan-Chiao (Barry) Cheng 鄭元僑
I'm a music AI researcher with 8 years of audio ML research and development experience in industry, and a guitarist and songwriter. I'm starting a Master in Music Technology at the Georgia Institute of Technology in Fall 2026.
About
I work on machine learning for music and audio. My background spans music information retrieval, representation learning, and audio effects modeling, with eight years of industry experience shipping audio ML to commercial products.
At Georgia Tech I'm broadening my research into music technology more generally: music generation, music understanding, and human and AI creative interaction. The thread I want to pull on is the same one that drew me in originally, which is building tools that meet musicians where they actually work.
Outside research, I write and play guitar in Diving Raccoon.
News
-
Paper accepted to ISMIR 2027
Our paper StemFX: Learning Mixing Style Representations via Autoregressive FX Chain Prediction on Source-Separated Stems was accepted to the International Society for Music Information Retrieval Conference (ISMIR) 2027. Demo page arXiv Source code
-
Joining Sony Computer Science Laboratories as a research intern
Working on music generation and zero-shot music editing at Sony Computer Science Laboratories (Sony CSL) under the supervision of Vincent Cheung. July–August 2026.
-
Invited as a reviewer for ICML 2026 Workshop on Machine Learning for Audio
Joining the review committee for the ICML 2026 Workshop on Machine Learning for Audio. View invitation
-
Two new singles & tour with Diving Raccoon
Releasing two singles and heading out on tour with my band Diving Raccoon. Dates and venues to be announced. Tour info
-
Invited as a reviewer for ISMIR 2026
Joining the review committee for the 27th International Society for Music Information Retrieval Conference, held in Abu Dhabi, UAE, November 8–12, 2026. View invitation
-
Accepted to Georgia Tech's Music Technology M.S. program
Admitted to the Music Technology Master's program at the Georgia Institute of Technology for the Fall 2026 semester. View admission letter
Experience
Work
-
Research Intern
Sony Computer Science Laboratories · Tokyo, Japan
Jul 2026 – Aug 2026
- Research on music generation and zero-shot music editing, under the supervision of Vincent Cheung.
-
Senior Machine Learning Engineer
Positive Grid · Taipei, Taiwan
Sep 2020 – Jun 2026
- Developed the foundational Tone Embedding model for electric guitar sound, the basis for advanced tone analysis and customization.
- Designed a zero-shot guitar amplifier modeling system that simulates diverse amplifiers without additional training data.
- Built a music source separation model that outperforms state-of-the-art methods on guitar tracks.
- Led Smart Jam, a real-time, user-interactive, symbolic-based music generation app, driving user engagement.
- Integrated ML models into mobile and desktop platforms; deployed 5+ models to production services.
-
Machine Learning Engineer
Intelligo Technology Inc. · Taipei, Taiwan
Apr 2018 – Jul 2020
- Developed speech enhancement and automatic speech recognition (ASR) models.
- Implemented audio ML models on embedded systems.
Reviewer
-
ICML 2026 Workshop on Machine Learning for Audio
Reviewer · 2026
-
International Society for Music Information Retrieval Conference (ISMIR) 2026
Reviewer · 2026
Education
-
M.S. Music Technology
Georgia Institute of Technology · Starting Fall 2026
-
M.S. Photonics and Optoelectronics
National Taiwan University · 2015 – 2017
-
B.S. Physics
National Taiwan University · 2011 – 2015
Music
I write and play guitar in Diving Raccoon. Listen on Spotify, and catch us on tour later this year.
Get in touch
For research, collaboration, or shows, drop me a line.