Minje Kim, Ph.D.

Bio: Minje Kim is an associate professor in the Siebel School of Computing and Data Science at the University of Illinois at Urbana-Champaign and an Amazon Visiting Academic working at Amazon Lab126. Before then, he was an associate professor at Indiana University (2016-2023). He earned his Ph.D. in Computer Science at UIUC (2016) after working as a researcher at ETRI, a national lab in Korea (2006 to 2011). Prior to that, he received his Master’s and Bachelor’s degrees in the Department of Computer Science and Engineering at POSTECH (Summa Cum Laude) and in the Division of Information and Computer Engineering at Ajou University (with honors) in 2006 and 2004, respectively. During his career as a researcher, he has focused on developing machine learning models for audio signal processing applications. He is a recipient of various awards, including the NSF Career Award (2021), IU Trustees Teaching Award (2021), IEEE SPS Best Paper Award (2020), Google and Starkey’s grants for outstanding student papers in ICASSP 2013 and 2014, respectively, and Richard T. Cheng Endowed Fellowship from UIUC in 2011. He is the Chair of the IEEE SPS Audio and Acoustic Signal Processing Technical Committee (2025-2026). He is serving as Senior Area Editor for IEEE/ACM Transactions on Audio, Speech, and Language Processing and IEEE Signal Processing Letters, Associate Editor for EURASIP Journal of Audio, Speech, and Music Processing, and Consulting Associate Editor for IEEE Open Journal of Signal Processing. He was the General Chair of IEEE WASPAA 2023 and also a reviewer, program committee member, or area chair for the major machine learning and signal processing venues. He is on more than 50 patents as an inventor.

My Students

Ph.D. Candidates

Haici Yang

Intro: Haici joined my group at IU in 2019. She has introduced various innovative ideas to neural coding systems, such as source separation and predictive coding conducted in the latent code space and the latent diffusion model for generative de-quantization. Other than neural coding, she also worked on music audio projects, such as end-to-end music source remixing networks and style transfer-based upmixing (from two to five channels). She interned at Amazon, Adobe, and MERL. Her dissertation research is about scalable diffusion models for neural speech/audio coding. Haici likes to learn musical instruments, loves outdoor activities, and reading books.

Homepage: https://haiciyang.github.io
Email: hy17-at-iu-edu
CV: PDF

Anastasia Kuznetsova

Intro: Anastasia is dual-majoring in computer science and computational linguistics at IU. Her research revolved around speech technology with a focus on self-supervised learning and low-resource languages. After joining my group, she broadened her interest to generative data augmentation for speech enhancement, foundational models for multichannel audio, and neural codecs (i.e., “tokenization”) for ASR. Anastasia interned at Coqui.ai, Rev.com, Google, and Amazon. Her dissertation is about improving data, resources, and communication efficiency in speech processing. She likes to lift heavy things, sing in a choir, and is into fashion.

Homepage: https://ana-kuznetsova.github.io/
Email: anakuzne at iu dot edu
CV: PDF

Darius Petermann

Intro: Darius came to my group at IU with his prior experiences in music technology. His main research topics have been innovating neural coding architectures for high-fidelity audio signals. Meanwhile, he has also established strong relationships with leading industry research labs throughout his internships at MERL, Google, and Netflix, one of which led to the Best Student Paper Award at ICASSP 2023. Besides neural coding, he has also worked heavily on spatial source separation problems. More recently, his research extended to multimodal and generative models. I like talking to him on heavy metal music, while he also likes to spend time with his dog Mötley.

Homepage: http://www.dariuspetermann.com
Email: daripete-at-indiana-dot-edu
CV: PDF

Ph.D. Students

Tsun-An Hsieh

Intro: Tsun-An joined my group in 2022 at IU, but then he moved with me to UIUC. He is knowledgeable in the ML literature, such as transfer learning, metric learning, and causal inference. With me, he started to focus on advancing speech separation using language models and the personalization concept.

Homepage: https://alexiehta.github.io
Email: tsunanh2-at-illinois-dot-edu
CV: HTML

Jaesung Bae
(co-advise with Paris Smaragdis)

Intro: A new student joined UIUC in 2024. More information is coming.

Homepage: https://jaesungbae.github.io/
Email: jb82-at-illinois-dot-edu
CV: PDF

Jackie Lin
(co-advise with Paris Smaragdis)

Intro: A new student joined UIUC in 2024. More information is coming.

Homepage:
Email:
CV:

Yutong Wen
(co-advise with Paris Smaragdis)

Intro: A new student joined UIUC in 2024. More information is coming.

Homepage:
Email:
CV:

Cameron Churchwell
(co-advise with Paris Smaragdis)

Intro: A new student joined UIUC in 2024. More information is coming.

Homepage:
Email:
CV:

Master Students

Jocelyn Xu

Intro: A new student joined UIUC in 2024. More information is coming.

Homepage:
Email:
CV:

Visiting Scholars

Jayeon Yi

Intro: An MS student at the U. of Michigan.

Homepage:
Email:
CV:

Alumni

Sanna Wager, Ph.D.
  • Ph.D. in Informatics (Spring 2021) and minor in Statistics
  • Dissertation: “A Data-Driven Pitch Correction Algorithm for Singing Voice” (pdf)
  • The committee: Minje Kim (chair), Christopher Raphael (IU CS), Donald Williamson (IU CS), and Daniel McDonald (U. of British Columbia Statistics)
  • Homepage: http://homes.sice.indiana.edu/scwager/
  • Work at SAIGE: Dr. Sanna Wager began to work with Minje in fall 2016, and joined the group officially in spring 2017. During her Ph.D. study, she wrote five conference papers on various research topics including tensor decomposition-based multhchannel speech dereverberation, robust speech recognition, semi-supervised methods for collecting a large-scale singing performance dataset, and neural pitch correction algorithms for singing voice. She interned at various companies, such as Google, Smule, Spotify, and Amazon. Her dissertation research on neural pitch correction algorithms for singing received extensive media coverage, including BBC, The Times, Daily Mail, the New Scientist Magazine, etc. She is now at Amazon as an Applied Scientist.
Kai Zhen, Ph.D.
  • Ph.D. (dual degree) in Computer Science and Cognitive Science (Spring 2021)
  • Dissertation: “Neural Waveform Coding: Scalability, Efficiency, and Psychoacoustic Calibration” (pdf)
  • The committee: Minje Kim (chair), Robert Goldstone (IU Cognitive Science), Donald Williamson (IU Computer Science), and Shen Yi (U. of Washington, Speech and Hearing Sciences)
  • Homepage: http://www.kaizhen.us
  • Work at SAIGE: Dr. Kai Zhen joined the lab in fall 2017 and led multiple neural waveform coding projects, which pioneered a new research area. His papers were published in leading signal processing and speech processing conferences and journals, such as Interspeech, ICASSP, IEEE Signal Processing Letters, and IEEE T-ASLP. The Cognitive Science Program at IU recognized his research by awarding the Outstanding Research Award in 2021. During his Ph.D. study, Dr. Zhen interned at LinkedIn (2018 and 2019) and Amazon (2020). During his internship at Amazon, his work on efficient neural ASR systems was selected as one of the 17 best poster presentations out of 180 internship projects. He is now at Amazon as an Applied Scientist.
Sunwoo Kim, Ph.D.
  • Ph.D. in Intelligent Systems Engineering (Spring 2022) and minor in Computer Science
  • Dissertation: “Model Compression for Efficient Machine Learning Inference” (pdf)
  • The committee: Minje Kim (chair), Peter Todd (IU Cognitive Science), Christopher Raphael (IU Computer Science), and Fan Chen (IU ISE)
  • Homepage: https://www.kimsunwoo.com
  • Work at SAIGE: Dr. Sunwoo Kim joined the lab in Fall 2017 and has focused on developing efficient machine learning models, such as bitwise neural networks, boosted hashing methods, and knowledge distillation. Out of various papers he published, his ICASSP 2020 paper was recognized as the Best Student Paper runner-up. He interned at Qualcomm (2019) and Amazon (2020 and 2021). He is now at Amazon as an Applied Scientist.
R. David Badger, Ph.D.
  • Ph.D. in Intelligent Systems Engineering (Spring 2022)
  • Dissertation: “Open-Source Classification Systems for Frequency-Domain RF Signals: Robust Physical Layer Multi-Sample Rate Processing” (pdf)
  • The committee: Minje Kim (chair), Lei Jiang (IU ISE), Lantao Liu (IU ISE), and Ariful Azad (IU ISE)
  • Work at SAIGE: Dr. R. David Badger joined the lab in Fall 2018. He has worked on RF signal processing, with a focus on signal compression and classification. He has open-sourced his dataset and deep learning-based RF classification system, which was the first of its kind that aims at multi-class RF signal classification using neural networks. He is now at Naval Surface Warfare Center Crane Division.
Aswin Sivaraman, Ph.D.
  • Ph.D. in Intelligent Systems Engineering (Spring 2024) and minor in Computer Science
  • Dissertation: “Resource-Efficient Model Adaptation Methods for Personalized Speech Enhancement Systems”
  • The committee: Minje Kim (chair), Christopher Raphael (IU Computer Science), David Crandall (IU Computer Science), and Ariful Azad (IU ISE)
  • Homepage: http://aswinsivaraman.com
  • Work at SAIGE: Dr. Aswin Sivaraman joined the lab in Fall 2017 and has focused on developing efficient machine learning models and data augmentation techniques, such as data purification, self-supervised learning from noisy signals, mixture of local experts, etc., for personalized speech enhancement. He wrote three conference papers (Interspeech and WASPAA) and one journal article (IEEE JSTSP) as the first author. He interned at Amazon, Spotify, Microsoft, and Google. He is now at Apple.
Mrinmoy Maity
  • MS in Intelligent Systems Engineering (Spring 2019)
  • Now at Microsoft as a Data Scientist