The Journal of the Acoustical Society of Korea (한국음향학회지)

The Acoustical Society of Korea (한국음향학회)
권호 표지
DOI QR코드

DOI QR Code

Speech enhancement system using the multi-band coherence function and spectral subtraction method

다중 주파수 밴드 간섭함수와 스펙트럼 차감법을 이용한 음성 향상 시스템

  • Oh, Inkyu ;
  • Lee, Insung (School of Information Communication Engineering, Chungbuk National University)
  • 오인규 (충북대학교 정보통신공학부) ;
  • 이인성 (충북대학교 정보통신공학부)
  • Received : 2019.02.25
  • Accepted : 2019.05.17
  • Published : 2019.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

This paper proposes a speech enhancement method through the process of combining the gain function with spectrum subtraction method in the two microphone array with close spacing. A speech enhancement method that uses a gain function estimated by the SNR (Signal-to Noise Ratio) based on the multi frequency band coherence function causes the performance degradation in high correlation between input noises of two channels. A new speech enhancement method is proposed where the weighted gain function is used by combining the gain function from the spectral subtraction. The performance evaluation of the proposed method was shown by comparison with PESQ (Perceptual Evaluation of Speech Quality) value which is an objective quality evaluation test provided by the ITU-T (International Telecommunications Union Telecommunication). In the PESQ tests, the maximum 0.217 of PESQ value is improved in the various background noise environments.

본 논문은 두 개의 마이크로폰 환경에서 다중 주파수 대역 이득함수와 주파수 차감법을 결합하여 배경잡음을 억제하는 방법을 제안하였다. 다중 주파수 대역 신호대잡음비 추정을 통해 이득 함수를 얻는 음성 향상 방법은 두 채널 간 잡음신호의 상관성이 큰 경우 잡음 제거 성능이 떨어지는 단점을 가지고 있다. 하나의 채널 에서 스펙트럼 차감법을 통해 얻은 이득함수와 간섭함수 기반의 신호대잡음비 추정을 통해서 얻은 이득함수를 결합하여 가중된 이득함수를 사용하는 음성 향상 방법을 제안하였다. 제안된 방법은 ITU-T(International Telecommunications Union Telecommunication)의 객관적인 품질 평가 방법인 PESQ(Perceptual Evaluation of Speech Quality) 시험과 스펙트로그램을 사용하여 성능 평가 되어졌고 PESQ시험에서 최대 MOS 0.217의 음질 향상을 얻을 수 있었다.

Keywords

GOHHBH_2019_v38n4_406_f0001.png 이미지

Fig. 1. Placement of two microphones and sound sources.

GOHHBH_2019_v38n4_406_f0002.png 이미지

Fig. 2. Noise suppression algorithm using multi-frequency band coherence function and spectral subtraction method (CSH-NS).

GOHHBH_2019_v38n4_406_f0003.png 이미지

Fig. 3. PESQ Performance evaluation of the proposed CSH-NS algorithm in a market buzz background noise.

GOHHBH_2019_v38n4_406_f0004.png 이미지

Fig. 4. PESQ Performance evaluation of the proposed CSH-NS algorithm in subway background noise.

GOHHBH_2019_v38n4_406_f0005.png 이미지

Fig. 5. PESQ Performance evaluation of the proposed CSH-NS algorithm in fire truck background noise.

GOHHBH_2019_v38n4_406_f0006.png 이미지

Fig. 6. PESQ Performance evaluation of the proposed CSH-NS algorithm in car background noise.

GOHHBH_2019_v38n4_406_f0007.png 이미지

Fig. 7. The spectrogam (a) original sound, (b) 16-band CF-NS algorithm in market buzz background noise, (c) proposed CSH-NS algorithm in market buzz background noise.

GOHHBH_2019_v38n4_406_f0008.png 이미지

Fig. 8. The spectrogam (a) original sound, (b) 16-band CF-NS algorithm in fire truck background noise, (c) proposed CSH-NS algorithm in fire truck background noise.

References

  1. M. Jelinek and R. Salami, "Noise reduction method for wideband speech coding," Proc. 12th European Signal Processing Conference, 1959-1962 (2004).
  2. M. Jeub, C. Herglots, C. Nelke, C. Beaugeant, and P. Vary, "Noise reduction for dual-microphone mobile phones exploiting power level differences," Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing(ICASSP), 1693-1696 (2012).
  3. C. M. Nelke, C. Beaugeant, and P. Vary, "Dual microphone noise PSD estimation for mobile phones in hands-free position exploiting the coherence and speech presence probability," Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing(ICASSP), 7279-7283 ( 2013).
  4. K. Lee, J. Cho, and Y. Park, "Channel predictionbased noise reduction algorithm for dual-microphone mobile phones," IEEE Transactions on Consumer Electronics, 393-491 (2014).
  5. N. Yousefian, K. Kokkinakis, and P. C. Loizou, "A coherence-based algorithm for noise reduction in dual-microphone application," Proc. Eur. Signal Process. Conf. (EUSIPCO'10), 1904-1908 (2010).
  6. I. Oh and I. Lee, "Multi frequency band noise suppression system using signal-to-noise ratio estimation", J. Acoust. Soc. Kr. 35, 102-109 (2016). https://doi.org/10.7776/ASK.2016.35.2.102
  7. T. V. Bogaert, S. Doclo, J. Wouters, and M. Moonen, "Speech enhancement with multichannel Wiener filter techniques in multi-microphone binaural hearing aids," J. Acoust Soc. Am. 360-371 (2009).
  8. R. Le Bouquin, A. Azirani, and G. Faucon, "Enhancement of speech degraded by coherent and inchoherent noise using a cross-spectral estimator," IEEE Trans. Speech Audio process, 484-487 (1997).
  9. R. Le. Bouquin and G. Faucon, "Using the coherence function for noise reduction," IEE Proc. I - Communications, Speech and Vision, 276-280 (1992).
  10. M. Brandstein and D. Ward, Microphone Arrays: Signal Processing Techniques and Applications (Springer-Verlag, Berlin, 2001), pp. 61-85.
  11. ITU-T Recommendation, Perceptual evaluation of speech quality (PESQ), an objective method for endto-end speech quality assessment of narrow-band telephone networks and speech codecs, ITU-T Recommendation, 2001.