Improving music source separation based on deep neural networks through data augmentation and network blending

Venue

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 261-265

Publication Year

2017

Keywords

Wiener filters;mixture models;music;recurrent neural nets;speech processing;SiSEC DSD100 dataset;blending scheme yields;data augmentation;deep neural network architectures;multichannel Wiener filter post-processing;music source separation improvement;recurrent network;Context;Indexes;Instruments;Recurrent neural networks;Source separation;Training;Blending;Deep neural network (DNN);Long-short term memory (LSTM);Music source separation (MSS)

Identifiers

Authors

  • S. Uhlich
  • M. Porcu
  • F. Giron
  • M. Enenkl
  • T. Kemp
  • N. Takahashi
  • Y. Mitsufuji

Source Materials