Journal of Broadcast Engineering (방송공학회논문지)

The Korean Institute of Broadcast and Media Engineers (한국방송∙미디어공학회)
권호 표지
DOI QR코드

DOI QR Code

Robust Multi-channel Wiener Filter for Suppressing Noise in Microphone Array Signal

마이크로폰 어레이 신호의 잡음 제거를 위한 강인한 다채널 위너 필터

  • Jung, Junyoung (School of Electrical Engineering, Soongsil University) ;
  • Kim, Gibak (School of Electrical Engineering, Soongsil University)
  • Received : 2018.05.08
  • Accepted : 2018.07.10
  • Published : 2018.07.30
Download PDF
⟨ Previous Next ⟩

Abstract

This paper deals with noise suppression of multi-channel data captured by microphone array using multi-channel Wiener filter. Multi-channel Wiener filter does not rely on information about the direction of the target speech and can be partitioned into an MVDR (Minimum Variance Distortionless Response) spatial filter and a single channel spectral filter. The acoustic transfer function between the single speech source and microphones can be estimated by subspace decomposition of multi-channel Wiener filter. The errors are incurred in the estimation of the acoustic transfer function due to the errors in the estimation of correlation matrices, which in turn results in speech distortion in the MVDR filter. To alleviate the speech distortion in the MVDR filter, diagonal loading is applied. In the experiments, database with seven microphones was used and MFCC distance was measured to demonstrate the effectiveness of the diagonal loading.

본 논문에서는 다채널 위너 필터를 이용하여 마이크로폰 어레이에서 취득된 신호의 잡음을 제거하는 방법을 다룬다. 다채널 위너 필터는 음성 신호의 방향에 대한 정보를 필요로 하지 않는 필터로서 단일 음성 음원의 경우, 음성 왜곡을 발생시키지 않는 MVDR (Minimum Variance Distortionless Response) 공간 필터와 단일 채널 스펙트럼 필터로 분리될 수 있다. MVDR의 방향벡터에 해당하는 단일 음성 음원과 마이크로폰 어레이 간의 음향 전달 함수는 다채널 위너 필터의 부공간 분해 (subspace decomposition)를 이용하여 추정할 수 있다. 이 때 상관 행렬 추정 과정에서 발생하는 오차로 인해 추정되는 음향 전달 함수에도 오차가 발생하게 되며 이에 따라 다채널 위너 필터를 구성하는 MVDR은 음성 왜곡을 발생시키게 된다. 이러한 음성 왜곡을 완화시키기 위해 diagonal loading을 적용하고 실험을 통해 그 효과를 검증한다. 실험에서는 7개의 선형 마이크로폰으로 수집된 데이터를 이용하였으며 잡음을 섞기 전 신호와 잡음을 섞은 후 필터를 통과시킨 신호 간의 MFCC 오차를 측정한다. 실험 결과, diagonal loading을 통해 MFCC 오차를 줄일 수 있음을 확인하였다.

Keywords

References

  1. M. Brandstein and D. Ward (Eds.), Microphone Arrays, SpringerVerlag, 2001. https://doi.org/10.1007/978-3-662-04619-7
  2. S. Doclo and M. Moonen, "GSVD-based optimal filtering for multi-microphone speech enhancement," IEEE Transactions on Signal Processing, vol. 50, no. 9, pp. 2230-2244, Sep. 2002. https://doi.org/10.1007/978-3-662-04619-7_6
  3. A. Spriet, M. Moonen, and J. Wouters, "Robustness Analysis of Multichannel Wiener Filtering and Generalized Sidelobe Cancellation for Multimicrophone Noise Reduction in Hearing Aid Applications," IEEE Transactions Speech Audio Processing, vol. 13, no. 4, July 2005, pp. 487-503. https://doi.org/10.1109/tsa.2005.845821
  4. S. Doclo and M. Moonen, "Combined Frequency-Domain Dereverberation and Noise Reduction Technique for Multi-microphone Speech Enhancement," Int. Workshop Acoustic Echo Noise Control, Darmstadt, Germany, Sept. 2001, pp. 31-34.
  5. W. Herbordt, Sound Capture for Human/Machine Interfaces, SpringerVerlag, 2005.
  6. M. Souden, J. Benesty, and S. Affes, "On optimal frequency-domain multichannel linear filtering for noise reduction," IEEE Transactions on Audio, Speech, and Language Processing, vol. 18, no. 2, pp. 260-276, 2010 https://doi.org/10.1109/tasl.2009.2025790
  7. Gibak Kim, "SNR-based weight control for the spatially preprocessed speech distortion weighted multi-channel Wiener filtering," Journal of Broadcast Engineering, Vol. 18, No.3, pp.455-462, March, 2013. https://doi.org/10.5909/jbe.2013.18.3.455
  8. R. Serizel, M. Moonen, B. Van Dijk, and J. Wouters, "Low-rank approximation based multichannel Wiener filter algorithms for noise reduction with application in cochlear implants," IEEE/ACM transactions on Audio, Speech, and Language Processing, vol. 22, no. 4, pp. 785-799, 2014. https://doi.org/10.1109/taslp.2014.2304240
  9. Z. Wang, "Rank-1 constrained Multichannel Wiener Filter for speech recognition in noisy environments," Computer Speech & Language, vol. 49, pp. 37-51, 2018. https://doi.org/10.1016/j.csl.2017.11.003
  10. D. Tylavsky and G. Sohie, "Generalization of the matrix inversion lemma," Proceedings of the IEEE, vol. 74, no. 7, pp. 1050-1052, 1986. https://doi.org/10.1109/proc.1986.13587
  11. G. Golub and C. Van Loan, Matrix Computations, Johns Hopkins University Press, 3rd edition, 1996. https://doi.org/10.2307/3619868
  12. H. Van Trees, Detection, Estimation and Modulation Theory, Part IV: Optimum Array Processing, New York: Wiley, 2002. https://doi.org/10.1002/0471221104
  13. R. Leonard, "A database for speaker-independent digit recognition," Proceedings of ICASSP, pp. 111-114, 1984. https://doi.org/10.1109/icassp.1984.1172716
  14. "RWCP Sound Scene Database in Real Acoustical Environments," Real World Computing Partnership, (C)1998-2001
  15. L. Du, J. Li, and P. Stoica, "Fully automatic computation of diagonal loading levels for robust adaptive beamforming," IEEE Transactions on Aerospace and Electronic Systems, vol. 46, no. 1, pp. 449-458, 2010. https://doi.org/10.1109/icassp.2008.4518112
  16. Y. Xiao, J. Yin, H. Qi, H. Yin, and G. Hua, "MVDR algorithm based on estimated diagonal loading for beamforming," Hindawi Mathematical Problems in Engineering, pp. 1-7, 2017. https://doi.org/10.1155/2017/7904356