LabCompass

Minje Kim

Computer Science · Indiana University

Mid career · publishing since 2002

Publications

171

Citations

2,700

Est. group size

Recurring co-author estimate

Active years

25

Publishing since 2002

Research summary
AI-generated

Minje Kim's research focuses on processing speech and audio signals using machine learning, including separating individual speakers from mixed audio, removing background noise, and compressing audio efficiently for transmission or storage. Much of the recent work explores generative AI methods (such as diffusion models) and techniques to make these systems smaller and faster to run on limited hardware. The work is largely published through computer science and signal processing venues, with some outputs shared as preprints before formal publication.

Speech separation and target speaker extractionSpeech enhancement and noise/reverberation removalAudio and speech coding (compression)Efficient and scalable neural network modelsGenerative models for audio (diffusion, flow matching)

Publication output has remained fairly steady over the past decade, with a peak around 2021 and consistent activity of roughly 11-18 papers per year in recent years.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026

Publication cadence
Publications per year over the last 10 years — averaging 11.8/year recently
2017: 9 publications172018: 10 publications182019: 14 publications192020: 13 publications202021: 26 publications26212022: 12 publications222023: 11 publications232024: 18 publications242025: 16 publications252026: 2 publications26
Publishes in
  • arXiv (Cornell University)×43
  • ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)×5
  • Cancer Research×5
  • Zenodo (CERN European Organization for Nuclear Research)×4
  • IEEE Journal of Selected Topics in Signal Processing×2
Similar researchers
By research-topic overlap

This profile was generated automatically from public scholarly data (OpenAlex). Group size and activity levels are estimates derived from co-authorship patterns.

Last updated Jul 19, 2026.

Claim or correct this profile