DocumentCode :
3233798
Title :
Multi-channel noise reduction with beamforming and masking-based Wiener filtering for human-robot interface
Author :
Hong, Jungpyo ; Cho, Keunseok ; Hahn, Minsoo ; Kim, Suhwan ; Jeong, Sangbae
Author_Institution :
Dept. of Electr. Eng., KAIST, Daejeon, South Korea
fYear :
2011
fDate :
6-8 Dec. 2011
Firstpage :
260
Lastpage :
264
Abstract :
In this paper, an efficient noise reduction algorithm is proposed for robust speech recognition. For the nonstationary noise reduction, frequency-domain beamforming-based speech enhancement is performed and masking-based Wiener filter is applied to the beamforming output. To design the masking-based Wiener filter, the spectrum of beamforming output is classified into noise spectrum and speech spectrum at each spectral bin by the inter-channel time delay between two reference inputs. Hamming windowing for the speech spectrum and noise spectrum is separately performed to smooth each spectrum. Then, the Wiener filtering is applied to the beamforming output. The performance of the proposed algorithm significantly improves the speech recognition accuracies and the signal-to-noise ratios.
Keywords :
Wiener filters; human-robot interaction; signal denoising; speech enhancement; speech recognition; Hamming windowing; frequency-domain beamforming; human-robot interface; masking-based Wiener filtering; multichannel noise reduction; noise spectrum; nonstationary noise reduction; robust speech recognition; speech enhancement; speech spectrum; Array signal processing; Delay effects; Noise; Speech; Speech recognition; Wiener filter; Speech enhancement; Wiener filter; beamforming; generalized sidelobe canceller;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Automation, Robotics and Applications (ICARA), 2011 5th International Conference on
Conference_Location :
Wellington
Print_ISBN :
978-1-4577-0329-4
Type :
conf
DOI :
10.1109/ICARA.2011.6144892
Filename :
6144892
Link To Document :
بازگشت