In ASR (Automatic Speech Recognition) applications, one of the most important issues in the real-time beamforming of microphone arrays is the inability to capture the whole acoustic dynamics via a finite-length of data and a finite number of array elements. For example, the reflected source signal impinging from the side-lobe direction presents a coherent interference, and the non-minimal phase channel dynamics may require an infinite amount of data in order to achieve perfect equalization (or inversion). All these factors appear as uncertainties or un-modeled dynamics in the receiving signals. Traditional adaptive algorithms such as NLMS that do not consider these errors will result in performance deterioration. In this paper, a time domain beamformer using H∞ filtering approach is proposed to adjust the beamforming parameters. Furthermore, this work also proposes a frequency domain approach called SPFDBB (Soft Penalty Frequency Domain Block Beamformer) using H∞ filtering approach that can reduce computational efforts and provide a purified data to the ASR application. Experimental results show that the adaptive H∞ filtering method is robust to the modeling errors and suppresses much more noise interference than that in the NLMS based method. Consequently, the correct rate of ASR is also enhanced.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Jwu-Sheng HU, Wei-Han LIU, Chieh-Cheng CHENG, "Robust Beamforming of Microphone Array Using H∞ Adaptive Filtering Technique" in IEICE TRANSACTIONS on Fundamentals,
vol. E89-A, no. 3, pp. 708-715, March 2006, doi: 10.1093/ietfec/e89-a.3.708.
Abstract: In ASR (Automatic Speech Recognition) applications, one of the most important issues in the real-time beamforming of microphone arrays is the inability to capture the whole acoustic dynamics via a finite-length of data and a finite number of array elements. For example, the reflected source signal impinging from the side-lobe direction presents a coherent interference, and the non-minimal phase channel dynamics may require an infinite amount of data in order to achieve perfect equalization (or inversion). All these factors appear as uncertainties or un-modeled dynamics in the receiving signals. Traditional adaptive algorithms such as NLMS that do not consider these errors will result in performance deterioration. In this paper, a time domain beamformer using H∞ filtering approach is proposed to adjust the beamforming parameters. Furthermore, this work also proposes a frequency domain approach called SPFDBB (Soft Penalty Frequency Domain Block Beamformer) using H∞ filtering approach that can reduce computational efforts and provide a purified data to the ASR application. Experimental results show that the adaptive H∞ filtering method is robust to the modeling errors and suppresses much more noise interference than that in the NLMS based method. Consequently, the correct rate of ASR is also enhanced.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e89-a.3.708/_p
Copy
@ARTICLE{e89-a_3_708,
author={Jwu-Sheng HU, Wei-Han LIU, Chieh-Cheng CHENG, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Robust Beamforming of Microphone Array Using H∞ Adaptive Filtering Technique},
year={2006},
volume={E89-A},
number={3},
pages={708-715},
abstract={In ASR (Automatic Speech Recognition) applications, one of the most important issues in the real-time beamforming of microphone arrays is the inability to capture the whole acoustic dynamics via a finite-length of data and a finite number of array elements. For example, the reflected source signal impinging from the side-lobe direction presents a coherent interference, and the non-minimal phase channel dynamics may require an infinite amount of data in order to achieve perfect equalization (or inversion). All these factors appear as uncertainties or un-modeled dynamics in the receiving signals. Traditional adaptive algorithms such as NLMS that do not consider these errors will result in performance deterioration. In this paper, a time domain beamformer using H∞ filtering approach is proposed to adjust the beamforming parameters. Furthermore, this work also proposes a frequency domain approach called SPFDBB (Soft Penalty Frequency Domain Block Beamformer) using H∞ filtering approach that can reduce computational efforts and provide a purified data to the ASR application. Experimental results show that the adaptive H∞ filtering method is robust to the modeling errors and suppresses much more noise interference than that in the NLMS based method. Consequently, the correct rate of ASR is also enhanced.},
keywords={},
doi={10.1093/ietfec/e89-a.3.708},
ISSN={1745-1337},
month={March},}
Copy
TY - JOUR
TI - Robust Beamforming of Microphone Array Using H∞ Adaptive Filtering Technique
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 708
EP - 715
AU - Jwu-Sheng HU
AU - Wei-Han LIU
AU - Chieh-Cheng CHENG
PY - 2006
DO - 10.1093/ietfec/e89-a.3.708
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E89-A
IS - 3
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - March 2006
AB - In ASR (Automatic Speech Recognition) applications, one of the most important issues in the real-time beamforming of microphone arrays is the inability to capture the whole acoustic dynamics via a finite-length of data and a finite number of array elements. For example, the reflected source signal impinging from the side-lobe direction presents a coherent interference, and the non-minimal phase channel dynamics may require an infinite amount of data in order to achieve perfect equalization (or inversion). All these factors appear as uncertainties or un-modeled dynamics in the receiving signals. Traditional adaptive algorithms such as NLMS that do not consider these errors will result in performance deterioration. In this paper, a time domain beamformer using H∞ filtering approach is proposed to adjust the beamforming parameters. Furthermore, this work also proposes a frequency domain approach called SPFDBB (Soft Penalty Frequency Domain Block Beamformer) using H∞ filtering approach that can reduce computational efforts and provide a purified data to the ASR application. Experimental results show that the adaptive H∞ filtering method is robust to the modeling errors and suppresses much more noise interference than that in the NLMS based method. Consequently, the correct rate of ASR is also enhanced.
ER -