Time-Frequency-Bin-Wise Beamformer Selection and Masking for Speech Enhancement in Underdetermined Noisy Scenarios. Yamaoka, K., Brendel, A., Ono, N., Makino, S., Buerger, M., Yamada, T., & Kellermann, W. In 2018 26th European Signal Processing Conference (EUSIPCO), pages 1582-1586, Sep., 2018.
Time-Frequency-Bin-Wise Beamformer Selection and Masking for Speech Enhancement in Underdetermined Noisy Scenarios [pdf]Paper  doi  abstract   bibtex   
In this paper, we present a speech enhancement method using two microphones for underdetermined situations. A conventional speech enhancement method for underdetermined situations is time-frequency masking, where speech is enhanced by multiplying zero or one to each time-frequency component appropriately. Extending this method, we switch multiple preconstructed beamformers at each time-frequency bin, each of which suppresses a particular interferer. This method can suppress an interferer even when both the target and an interferer are simultaneously active at a given time-frequency bin. As a switching criterion, selection of minimum value of the outputs of the all beamformers at each time-frequency bin is investigated. Additionally, another method using direction of arrival estimation is also investigated. In experiments, we confirmed that the proposed methods were superior to conventional time-frequency masking and fixed beamforming in the performance of speech enhancement.
@InProceedings{8553299,
  author = {K. Yamaoka and A. Brendel and N. Ono and S. Makino and M. Buerger and T. Yamada and W. Kellermann},
  booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
  title = {Time-Frequency-Bin-Wise Beamformer Selection and Masking for Speech Enhancement in Underdetermined Noisy Scenarios},
  year = {2018},
  pages = {1582-1586},
  abstract = {In this paper, we present a speech enhancement method using two microphones for underdetermined situations. A conventional speech enhancement method for underdetermined situations is time-frequency masking, where speech is enhanced by multiplying zero or one to each time-frequency component appropriately. Extending this method, we switch multiple preconstructed beamformers at each time-frequency bin, each of which suppresses a particular interferer. This method can suppress an interferer even when both the target and an interferer are simultaneously active at a given time-frequency bin. As a switching criterion, selection of minimum value of the outputs of the all beamformers at each time-frequency bin is investigated. Additionally, another method using direction of arrival estimation is also investigated. In experiments, we confirmed that the proposed methods were superior to conventional time-frequency masking and fixed beamforming in the performance of speech enhancement.},
  keywords = {array signal processing;direction-of-arrival estimation;interference suppression;speech enhancement;time-frequency analysis;speech enhancement method;interference suppression;time-frequency-bin-wise beamformer masking;microphones;direction of arrival estimation;multiple preconstructed beamformers;underdetermined noisy scenarios;time-frequency-bin-wise beamformer selection;Time-frequency analysis;Direction-of-arrival estimation;Speech enhancement;Microphones;Estimation;Array signal processing;Distortion},
  doi = {10.23919/EUSIPCO.2018.8553299},
  issn = {2076-1465},
  month = {Sep.},
  url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439259.pdf},
}
Downloads: 0