한국음향학회지 (The Journal of the Acoustical Society of Korea)

  • 제24권7호
  • /
  • Pages.423-430
  • /
  • 2005
  • /
  • 1225-4428(pISSN)
  • /
  • 2287-3775(eISSN)
한국음향학회 (The Acoustical Society of Korea)
권호 표지

차량환경에서 음성인식 성능 향상을 위한 마이크로폰 어레이 빔형성 기법

A Microphone Array Beamformer for the Performance Enhancement of Speech Recognizer in Car

  • 한철희 (연세대학교 전기전자공학과) ;
  • 강홍구 (연세대학교 전기전자공학과) ;
  • 황영수 (관동대학교 전자정보통신기술공학부) ;
  • 윤대희 (연세대학교 전기전자공학과)
  • 발행 : 2005.10.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 차량환경에서 잔향과 근접장 효과에 의해 발생하는 목적 음성 신호의 왜곡을 감소시킬 수 있는 마이크로폰 어레이 빔형성 기법을 제안하였다. 온라인으로 추정하기 어려운 소스와 마이크간의 전달함수 대신 상대적으로 추정이 용이한 기준 마이크와 다른 마이크간의 상대전달함수를 조향 벡터로 이용함으로써, 원격장 모델의 조향 벡터를 이용한 빔형성기에 비해 목적 음성 신호의 왜곡을 감소시킬 수 있는 준최적 빔형성 기법을 제안하였다. 제안된 방법의 성능을 검증하기 위해, 실제 차량에서 녹음된 음성 DB를 구축하고, 이를 이용하여 HTK를 통한 음성인식 실험을 수행하였다. 음성인식 실험 결과 원격장 모델을 이용한 방법보다 인식률이 최대 $15\%$까지 향상됨을 확인하였다.

In this paper. a microphone array beamforming algorithm that reduces the signal distortion caused by reverberation and near-field effect in car environment is proposed. When reverberation or near-field effect is present, an optimum beamformer should be constructed with a steering vector consisting of transfer functions between source and microphones, but it is generally difficult to estimate transfer functions on-line without knowledge of the source signal. Instead, a sub-optimal beamforming algorithm that reduces signal distortion is proposed. It is constructed with steering vectors consisting of relative transfer functions between reference sensor and other sensors. In order to evaluate the performance of the proposed algorithm. we had recorded noisy speech database in a car, and performed speech recognition experiments with HMM Toolkit (HTK) released by Cambridge University. The recognition rate of the proposed algorithm was 15 percents higher than that of the conventional far-field beamformers in best case.

키워드

참고문헌

  1. M. Bransdstein and D. Ward, Microphone Arrays, (Springer, 2001)
  2. H. L. Van Trees, Optimum Arrav Processing, (Wiley-Interscience, 2002)
  3. S. Gannot, D Burnshtein, and E. Weinstein, 'Signal enhancement using beamforming and nonstationarity with applications to speech', IEEE Trans. Signal Processing, 49, 1614-1626, Aug. 2001 https://doi.org/10.1109/78.934132
  4. F. Asano, H. Aso, and T. Matsui, 'Sound Source Localization and Separation in Near Field', IEICE Trans. Fundamentals, E38-A, 2286-2294, Nov. 2000
  5. F. Asano et al., 'Real-time sound source localization and separation system and its application to automatic speech recognition', in Proc. EUROSPEECH2001, Aalborg, Denmark, Sept. 2001, 1013-1016
  6. S. Affes and Y. Grenier, 'A signal subspace tracking algorithm for microphone array processing of speech', IEEE Trans. Speech and Audio Processing, 5, 425-437, Sept. 1997 https://doi.org/10.1109/89.622565
  7. O. L. Frost, III, 'An algorithm for linearly constrained adaptive array processing', Proc. IEEE, 60, 926-935, Jan. 1972
  8. L. J. Griffiths and C. W. Jim, 'An alternative approach to linearly constrained adaptive beamforming', IEEE Trans. Antennas and Propagations, AP-30, 27-34, Jan. 1982
  9. http://htk.eng.cam.ac.uk/
  10. O. Shalvi and E. Weinstein, 'System identification using nonstationarv signals', IEEE Trans. Signal Processing, 44, 2055-2063, Aug. 1996 https://doi.org/10.1109/78.533725
  11. I. Cohen, 'Multichannel Post-Filtering in Nonstationarv Noise Environments', IEEE Trans. Signal Processing, 52,1249-1150, May 2004
  12. L. R. Rabiner and R. W. Schafer, Digital Processing of Speech Signals, (Prentice-Hall, 1978)
  13. I. Cohen, 'Relative transfer function identification using speech signals', IEEE Trans. Speech and Audio Processing, 12, 451-459, Sept. 2004 https://doi.org/10.1109/TSA.2004.832975
  14. R. E. Crochiere and L. R. Rabiner, Multirate Digital Signal Processing, (Prentice-Hall, 1983)