A Study of Transfer Learning in Music Source Separation
June 2020 - October 2020
I studied the effect of transfer learning on music source separation on domains with small amounts of data, circumventing the need to have massive labeled datasets for a target domain.
nussl: Northewestern University Source Separation Library
March 2020 - Present
nussl is a python package for quick and easy implementation of audio source separation. I specifically contributed utilities for data augmentation and inferfacing with datasets.
NylonRNN
November 2019 - December 2019
A deep learning system to generate segments of classical guitar pieces. The final project for COMP_SCI 397 Computational Creativity.