About Me
I am an instrument maker.
While traditional makers have used wood, metal, circuits, or rule-based software synthesizers,
I create instruments with AI.
Currently, I work on generative music models at Neutune for the MixAudio service.
During my PhD, I focused on DJ mix analysis and music structure analysis.
Web Demonstrations
DJ StructFreak 🤖
AI DJ Demo
Music Dissector 🔪
Interactive Music Structure Visualizer
DJ Mixer Analyzer 🎛️
Explains How DJ Controlled Mixer
Publications
All-In-One Metrical And Functional Structure Analysis With Neighborhood Attentions on Demixed Audio
Taejun Kim, and Juhan Nam
IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), 2023
[ pdf ] [ python package ] [ ai dj demo ] [ visual demo ] [ hugging face space ]
*** Best Student Paper Award ***
Temporal Feedback Convolutional Recurrent Neural Networks for Speech Command Recognition
Taejun Kim, and Juhan Nam
The Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), 2022
Joint Estimation of Fader and Equalizer Gains of DJ Mixers using Convex Optimization
Taejun Kim, Yi-Hsuan Yang, and Juhan Nam
The International Conference on Digital Audio Effects (DAFx), 2022
Reverse-Engineering The Transition Regions of Real-World DJ Mixes using Sub-band Analysis with Convex Optimization
Taejun Kim, Yi-Hsuan Yang, and Juhan Nam
The International Conference on New Interfaces for Musical Expression (NIME), 2021
A Computational Analysis of Real-World DJ Mixes using Mix-To-Track Subsequence Alignment
Taejun Kim, Minsuk Choi, Evan Sacks, Yi-Hsuan Yang, and Juhan Nam
The International Society for Music Information Retrieval Conference (ISMIR), 2020
Comparison and Analysis of SampleCNN Architectures for Audio Classification
Taejun Kim, Jongpil Lee, and Juhan Nam
IEEE Journal of Selected Topics in Signal Processing, 2019
Sample-level CNN Architectures for Music Auto-tagging Using Raw Waveforms
Taejun Kim, Jongpil Lee, and Juhan Nam
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018
Raw Waveform-based Audio Classification Using Sample-level CNN Architectures
Jongpil Lee, Taejun Kim, Jiyoung Park, and Juhan Nam
Machine Learning for Audio Signal Processing Workshop, Neural Information Processing Systems (NIPS), 2017
[ pdf ]
Education
Korea Advanced Institute of Science and Technology (KAIST)
Ph.D in Graduate School of Culture Technology (Music Information Retrieval)
2019 - 2024
University of Seoul
M.S. in Electrical and Computer Engineering
2017 - 2019
@ Data Mining Lab (Advisor: Hanjoon Kim)
University of Seoul
B.S. in Electrical and Computer Engineering
2012 - 2017
Research Interests
Generative Music Models, AI DJ, Music Information Retrieval, Machine Learning, Signal Processing, Musical Applications
Skills
Machine Learning
PyTorch, Scikit‐learn, Weights & Biases, TensorFlow
Data Engineering
Crawling, Elasticsearch, MongoDB, AWS Batch, Apache Spark
Data Analysis
Pandas, Numpy, Matplotlib, SQL, Jupyter Lab, Apache Zeppelin
Signal Processing
Librosa, Madmom, SciPy
Cloud Engineering
(Experienced in AWS) AWS Batch, Step Functions, Lambda, Cognito, S3, Elastic Beanstalk, API Gateway, Cloudfront, etc.
Web Development
Vue.js, Flask, Node.js, AWS Amplify
Programming
Python, JavaScript, Scala, SQL, R, Java
Musical Performance
Vinyl & Digital DJ, Electric Bass Guitar, Acoustic Fingerstyle Guitar