Keywords: Music Information Retrieval, Machine learning, deep learning
Career
- Mar 2022 – Now | AI Scientist/Director, Gaudio Lab, New York City/Seoul
- Aug 2020 – Mar 2022 | Senior Research Scientist, ByteDance AI Research Lab, Mountain View, California, USA
- July 2018 – Aug 2020 | Research Scientist at Spotify (MIQ), New York, USA
Education
- Oct 2014 – March 2018 | PhD Student at c4dm, EECS, Queen Mary University of London, UK
(@c4dm, supervisors: Mark Sandler and George Fazekas) - 2009 – 2011 | M.S. Electric Engineering and Computer Science at Applied Acoustics Lab, Seoul National University, Korea
(Digital audio signal processing, acoustic system modelling, sound field reproduction) - 2005 – 2009 | B.S. Electric Engineering and Computer Science at Seoul National University, Korea
Experiences
- 2016 summer/fall | Visiting PhD at New York University (@Center for Data Science, with Kyunghyun Cho)
- 2015 summer | Research Intern at Naver Labs, Naver Corporation, Korea
- 2011 – 2014 | Researcher at ETRI (Electronics and Telecommunications Research Institute), Korea
Publications
Please check out the publication page
Codes
- Kapre: Keras Audio Preprocessor | Documentation
- Transfer learning for music (convnet feature extractor)
- LSTM realbook, LSTMetallica
Reviews
Journals – IEEE Transactions on Audio, Speech, and Language Processing, IEEE Transactions on Multimedia, EURASIP, Neural Computing and Applications, IEEE Journal of Selected Topics in Signal Processing, Journal of Audio Engineering Society, IEEE Transactions on Signal Processing
Conferences – WAC 2017, ISMIR 2017, CMMR 2017, AAAI 2017, NIPS 2016 Workshop, ISMIR 2018, ICML 2019, ICASSP 2019, ISMIR2019, WSDM 2019 Workshop, CSMC 2019, NeurIPS 2019, ICLR 2020, ICLR 2020, NeurIPS 2020, AAAI 2020, etc.
Scholarships and awards
- ISMIR 2017 Best paper award
- EPSRC funding covering my tuition fee and stipend for my phd on FAST project (2014-2018)
- National Science & Technology Scholarship, 2005-2008 (Fully funded, Supported by the Korea Student Aid Foundation (KOSAF) grant funded by the Korea government)
Profiles
한글 블로그, Github, twitter, LinkedIn, Google Scholar
I don’t have access to my QMUL email from Jan 2019. Please use my Spotify email address, which you can find from my recent papers like this.
Others
I made some music. Some of them are on my Soundcloud.