Improving Music Source Separation Based on Deep Neural Networks through Data Augmentation and Network Blending


Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 261–265

Publication Year


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



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

Source Materials