DOI QR코드

DOI QR Code

Two-Channel Noise Reduction Using Beamforming and DOA-Based Masking

빔포밍 및 DOA 기반의 마스킹을 이용한 2채널 잡음제거

  • Received : 2012.07.30
  • Accepted : 2012.08.17
  • Published : 2013.01.31

Abstract

In this paper, we propose a multi-channel speech enhancement algorithm using beamforming and direction-of-arrival (DOA)-based masking. The proposed algorithm enhances noisy speech basically by the linearly constrained minimum variance (LCMV) algorithm and then a mel-scale Wiener filter designed using DOA-based masking is applied to remove still remaining noises. To improve the performance, we optimize the learning rate of the adaptive filters in LCMV and the DOA threshold to detect target speech spectrum. As performance indices, the perceptual evaluation of speech quality (PESQ) score and output SNRs are measured. Experimantal results show that the proposed algorithm outperforms the conventional LCMV beamformer by 0.09 in PESQ score and 5.75 dB in output SNR, respectively.

본 논문에서는 빔포밍과 입사각분석 기반 마스킹을 이용한 다채널 음성개선 알고리즘이 제안된다. 제안된 알고리즘에서는 LCMV 빔포밍을 수행한 후에 입사각 분석을 이용한 멜-주파수 위너필터가 적용되어 잔존하는 잡음을 제거한다. 성능 향상을 위해서 빔포밍의 적응 필터 학습률과 목표 음성 스펙트럼 검출을 위한 입사각 임계치가 최적화된다. 성능 지수로서 PESQ와 출력 SNR이 측정되었으며 실험 결과 제안한 알고리즘이 종전의 최소분산 빔포밍 기법보다 PESQ 관점에서 0.09, 출력 SNR 관점에서 5.75 dB의 성능 향상시킴을 알 수 있었다.

Keywords

References

  1. S. Jeong andM. Hahn, "Speech quality and recognition rate improvement in car noise environments," Electronics Letters, vol. 37, no. 12, pp. 801-802, 2001.
  2. ES 202 212 V1.1.2 "Speech processing, transmission and quality aspects(STQ); distributed speech recognition; extended advanced front-end feature extraction algorithm; compression algorithm; back-end speech reconstruction algorithm," ETSI Standard, 2005.
  3. B. D. Van Veen and K.M. Buckley, "Beamforming: A versatile approach to spatial filtering", IEEE ASSP Magazine, vol. 5, no. 2, pp. 4-24, 1998.
  4. M. Brandstein and D. Ward, Microphone Arrays: Signal Processing Techniques and Applications, Springer, 2001.
  5. J. Benesty, J. Chen, and Y. Huang, Microphone Array Signal Processing (Springer Topics in Signal Processing), Springer, 2008.
  6. A. Hyvarinen, and E. Oja, "Independent component analysis: Algorithms and applications," Neural Networks, vol. 13, no. 4, pp. 411-430, 2000. https://doi.org/10.1016/S0893-6080(00)00026-5
  7. 이영재, 김수환, 한승호, 한민수, 김영일, 정상배, "확률적 목표 음성 검출을 통한 다채널 입력 기반 음성개선," 한국음성학회학술지 말소리와 음성과학, 1권, 3호, pp. 97-104, 2009.
  8. 박지훈,이성주,홍정표,정상배,한민수(2008). "필 터뱅크 기반 프로스트 알고리즘을 이용한 빔포밍 최적화," 대한음성학회 학술지 말소리, 66호, pp. 73-86, 2008.
  9. L. Wang, H. Ding, and F. Yin, "Combining superdirective beamforming and frequency-domain blind source separation for highly reverberant signals," EURASIP Journal on Audio, Speech, and Music Processing, vol. 2010, pp. 1-13, 2010.
  10. O. L. Frost, "An algorithm for linearly constrained adaptive array processing," Proceedings of the IEEE, vol. 60, no. 8, pp. 926-935, 1972. https://doi.org/10.1109/PROC.1972.8817
  11. S. Jeong, H. Yang, and M. Hahn, "Two-channel noise reduction for robust speech recognition in car environments,"Electronics Letters, vol. 44, no. 17, pp. 1042-1043, 2008. https://doi.org/10.1049/el:20081811
  12. S. Jeong, S. Lee, and M. Hahn, "Dual microphonebased speech enhancement by spectral classification and Wiener filtering," Electronics Letters, vol. 44, no. 3, pp. 253-254, 2008. https://doi.org/10.1049/el:20083327
  13. 김수환,이영재,김영일,정상배, "DOA 기반학습률 조절을 이용한 다채널 음성 개선 알고리즘,"한국음성학회 학술지 말소리와 음성과학, 3권, 3호, pp. 91-98, 2011.
  14. http://en.wikipedia.org/wiki/PESQ