Improving DNN-based Music Source Separation using Phase Features

2018/07/07 by Joachim Muth, Muth, Joachim, Stefan Uhlich +10 · 1 citation
Computer Science · Engineering · #Blind Source Separation Techniques #Music and Audio Processing #Speech and Audio Processing #cs.LG #cs.SD #eess.AS

paper · pdf · doi:10.48550/arxiv.1807.02710

7 pages, 9 figures, Joint Workshop on Machine Learning for Music at ICML, IJCAI/ECAI and AAMAS, 2018

arxiv created 2018/07/16 · arxiv updated 2018/07/17

Abstract

Music source separation with deep neural networks typically relies only on amplitude features. In this paper we show that additional phase features can improve the separation performance. Using the theoretical relationship between STFT phase and amplitude, we conjecture that derivatives of the phase are a good feature representation opposed to the raw phase. We verify this conjecture experimentally and propose a new DNN architecture which combines amplitude and phase. This joint approach achieves a better signal-to distortion ratio on the DSD100 dataset for all instruments compared to a network that uses only amplitude features. Especially, the bass instrument benefits from the phase information.

Cited by

• Deep Learning Based Phase Reconstruction for Speaker Separation: A Trigonometric Perspective

Related

• Does Phase Matter For Monaural Source Separation?
• End-to-end music source separation: is it possible in the waveform domain?
• All for One and One for All: Improving Music Separation by Bridging Networks
• Phase reconstruction based on recurrent phase unwrapping with deep neural networks
• Blind Source Separation in Polyphonic Music Recordings Using Deep Neural\n Networks Trained via Policy Gradients