Blind Source Separation Based on a Beamformer Array and Time Frequency Binary Masking

Jan Cermak, Shoko Araki, Hiroshi Sawada, Shoji Makino

Abstract

This paper deals with a new technique for blind source separation (BSS) from convolutive mixtures. We present a three-stage separation system employing time-frequency binary masking, beamforming and a non-linear post processing technique. The experiments show that this system outperforms conventional time-frequency binary masking (TFBM) in both (over-)determined and underdetermined cases. Moreover it removes the musical noise and reduces interference in time-frequency slots extracted by TFBM.
BibTeX
@inproceedings{cermak2007blind,
  author = {Jan Cermak and Shoko Araki and Hiroshi Sawada and Shoji Makino},
  title = {{Blind Source Separation Based on a Beamformer Array and Time Frequency Binary Masking}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2007},
  doi = {10.1109/icassp.2007.366637},
}

Loading related content...

Figure