MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals

@article{Sawada2007MLSP2D,
  title={MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals},
  author={Hiroshi Sawada and Shoko Araki and Shoji Makino},
  journal={2007 IEEE Workshop on Machine Learning for Signal Processing},
  year={2007},
  pages={45-50}
}
This paper describes the frequency-domain approach to the blind source separation of speech/audio signals that are convolutively mixed in a real room environment. With the application of short- time Fourier transforms, convolutive mixtures in the time domain can be approximated as multiple instantaneous mixtures in the frequency domain. We employ complex-valued independent component analysis (ICA) to separate the mixtures in each frequency bin. Then, the permutation ambiguity of the ICA… CONTINUE READING