Reference-Invariant Training for Neural Array-Generic DOA Estimation
Jackie Lin, Saurabhchand Bhati, Anton Ratnarajah, Mrudula Athi, Wontak Kim, Minje Kim — Under review at ICASSP 2027
Jackie Lin
My research develops machine learning methods for room acoustics, spatial audio, and generative audio, with an emphasis on estimating, completing, and controllably generating room impulse responses.
About
Hi I'm Jackie! I'm a doctoral candidate in the Computer Science department at the University of Illinois Urbana-Champaign (UIUC), advised by Professor Minje Kim. I am interested in machine learning for room acoustics, spatial audio, and generative audio.
I received a Master's degree in Acoustics and Audio Technology from Aalto University in Espoo, Finland, advised by Professor Sebastian J. Schlecht. There, I worked on a deep learning system for estimating late reverberation and blind reverberation estimation.
Before starting my audio journey, I got my undergraduate degree in architecture and urban planning at MIT, and worked in Colorado as an architect. Check out my Projects page to see what I've been working on recently and some foundational projects. :)
Publications
Under review
2026
Few-Shot Room Impulse Response Interpolation in Latent Domains
Jackie Lin, Yoshiki Masuyama, Christoph Boeddeker, Julius Richter, Gordon Wichern, Minje Kim, Jonathan Le Roux — 2026 19th International Workshop on Acoustic Signal Enhancement (IWAENC)
Gencho: Room Impulse Response Generation from Reverberant Speech and Text via Diffusion Transformers
Jackie Lin, Jiaqi Su, Nishit Anand, Zeyu Jin, Minje Kim, Paris Smaragdis — ICASSP 2026
2025
Deep Room Impulse Response Completion
Jackie Lin, Georg Götz, Sebastian J. Schlecht — EURASIP Journal on Audio, Speech, and Music Processing, 2025
Generative Data Augmentation Challenge: Synthesis of Room Acoustics for Speaker Distance Estimation
Jackie Lin, Georg Götz, Hermes Sampedro Llopis, Haukur Hafsteinsson, Steinar Guðjónsson, Daniel Gert Nielsen, Finnur Pind, Paris Smaragdis, Dinesh Manocha, John Hershey, Trausti Kristjansson, Minje Kim — GenDA Workshop @ ICASSP 2025
2023
Deep Learning with Audio: An Explorative Syllabus for Music Composition and Production
Koray Tahiroğlu, Shenran Wang, Eduard Tampu, Jackie Lin — International Conference on AI and Musical Creativity (AIMC), 2023
Experience
Shure Inc.
May – Aug 2026Digital Signal Processing Intern with the Signal Processing and Applied Research team · Hosts: Dr. Mark Syordenko, Dr. Wenyu Jin
- Triaged failure modes of a spatial source separation algorithm (SSSA) for a large, conferencing microphone array.
- Developed a deep-learning room geometry estimator using room impulse response (RIR) inputs.
- Developed a speech quality estimator for the SSSA given room acoustics and room-device-source spatial information.
Mitsubishi Electric Research Laboratories (MERL)
2025 – 2026Research · Hosts: Dr. Yoshiki Masuyama, Dr. Gordon Wichern, Dr. Jonathan Le Roux
- Neural spatial audio processing and understanding systems
- Generative AI for audio
Adobe Research
2025Research · Hosts: Dr. Jiaqi Su, Dr. Zeyu Jin
- Diffusion transformer–based blind RIR generation (Gencho)
- Text-controllable RIR generation
University of Illinois Urbana-Champaign
2024 – PresentGraduate Research · Advisors: Prof. Paris Smaragdis, Prof. Minje Kim · Urbana, IL
- Spatial audio understanding and generation
- Audio-visual scene understanding
- Neural representations for RIRs and acoustic scenes
- Generative data augmentation for high-quality RIRs and downstream tasks
Aani Oy
2024Research · Supervisors: Dr. Raimundo Gonzalez Diaz, Prof. Archontis Politis
- Deep learning for blind reverberation estimation
- PyTorch pipelines for new blind reverberation systems
Aalto University — Virtual Acoustics
2023 – 2024Master’s thesis research · Supervisors: Prof. Sebastian J. Schlecht, Prof. Vesa Välimäki, Georg Götz
- Encoder–decoder networks for RIR completion from early reflections
- Lightweight decoder for late-reverberation synthesis; trained on 50k+ RIRs (Triton cluster)
Bose Corporation
2023Acoustic Test Engineering Intern · Framingham, MA
- MVDR beamforming and acoustic test parameters for noise rejection
- MATLAB library for multichannel denoising from variable microphone arrays
Education
University of Illinois Urbana-Champaign
2024 – PresentPh.D., Computer Science · Urbana, Illinois
Aalto University
2022 – 2024M.Sc., Acoustics and Audio Technology (Honors); Minor in Machine Learning · Espoo, Finland
- Fully funded M.Sc. tuition award
Massachusetts Institute of Technology
2016 – 2020B.S., Architecture and Urban Planning · Cambridge, Massachusetts
Skills
- Programming & machine learning
- Python, PyTorch, scikit-learn, SLURM, Git
- Acoustics & audio
- MATLAB, audio DSP, room and loudspeaker measurement
- Spatial / interactive audio
- Unity, Unreal Engine, Oculus development, game audio (Wwise)
Contact
Reach me through the UIUC email link or on GitHub. Download the CV (PDF), and see selected work on the Projects page.