Jongho Choi

MS Student (2020~)

MARG, Seoul National University

Deep Learning Research Engineer

Broad and shallow knowledge could be a good weight initialization for various problems.

Interests

• Deep Learning
• Automatic Music Transcription
• Automatic Piano Arrangement
• Music Recommendation

Education

• MS student in Intelligence and Information, 2020~

  Graduate School of Convergence Science and Technology, Seoul National University

• BS in Electrical and Computer Engineering, 2013~2020

  Dept. of Engineering, Seoul National University

Experience

 

 

 

 

 

Software Engineer (As Substitutional Military Service)

GemizipSoft

Jul 2016 – Sep 2017 Seoul

Real-time Derivative Trading Data Feeding System

 

 

 

 

 

Research Engineer (As Substitutional Military Service)

ESTsoft

Sep 2017 – Jan 2020 Seoul

• DL-based Eyeware Image Retrieval System
• 3D AR Eyeware Rendering Library

 

 

 

 

 

Research Intern

Kakao Corp

Jun 2020 – Jul 2020 SeongNam

Recommendation Team(Kakao RecoTeam)

Projects

*

All Open Source Other

Cerberus

Implementation of Simultaneous Separation and Transcription of Mixtures with Multiple Polyphonic and Percussive Instruments (Ethan Manilow et al., ICASSP2020)

Video Github

DDSP : Differentiable DSP

Implementation of DDSP (PyTorch), Differentiable Digital Signal Processing (ICLR 2020)

Demo Github

Deep Complex U-Net

PyTorch Implementation of Phase-Aware Speech Enhancement with Deep Complex U-Net, (H. Choi et al., 2018)

Github

Onsets and Frames

Implementation of Onsets and frames: Dual-objective piano transcription (Curtis Hawthorne et al., ISMIR 2018)

Demo