Resource Efficient Deep Eigenvector Beamforming

Matthias Zöhrer; Lukas Pfeifenberger; Günther Schindler; Holger Fröning; Franz Pernkopf

ICASSP 2018accepted0 citations

Resource Efficient Deep Eigenvector Beamforming

Matthias Zöhrer, Lukas Pfeifenberger, Günther Schindler, Holger Fröning, Franz Pernkopf

Abstract

We propose binary neural networks (BNN s) for acoustic beamforming. This makes the speech enhancement approach resource efficient and applicable for embedded applications. Using CHiME4 data, we use BNN s to estimate the speech presence probability mask for GEV-PAN beamformers. By doing so, we achieve audio quality and ASR scores on par to single-precision deep neural networks (DNNs), while the computational requirements and the memory footprint are significantly reduced.

BibTeX

@inproceedings{icassp2018_resourceefficien,
  title = {Resource Efficient Deep Eigenvector Beamforming},
  author = {Matthias Zöhrer and Lukas Pfeifenberger and Günther Schindler and Holger Fröning and Franz Pernkopf},
  booktitle = {ICASSP 2018},
  year = {2018}
}