한국전자통신학회논문지 (The Journal of the Korea institute of electronic communication sciences)

한국전자통신학회 (Korea Institute of Electronic Communication Science)
권호 표지
DOI QR코드

DOI QR Code

음원신호 추출을 위한 주파수영역 응용모델에 기초한 독립성분분석

Independent Component Analysis Based on Frequency Domain Approach Model for Speech Source Signal Extraction

  • 최재승 (신라대학교 스마트전기전자공학부)
  • Choi, Jae-Seung (Division of Smart Electrical and Electronic Engineering, Silla University)
  • 투고 : 2020.07.13
  • 심사 : 2020.10.15
  • 발행 : 2020.10.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문은 여러 음원신호가 혼합된 환경에서 목적으로 하는 음원신호만을 분리하기 위하여 마이크로폰을 사용한 블라인드 음원분리 알고리즘을 제안한다. 제안하는 알고리즘은 독립성분분석 방법을 기반으로 한 주파수영역 표현모델이다. 따라서 2 음원에 대한 주파수영역 독립성분분석의 실제 환경에서의 유효성 검증을 목적으로, 음원의 종류를 변경하여 주파수영역 독립성분분석을 실행하여 음원분리를 실시하여 그 향상효과를 검증한다. 파형에 의한 실험결과로부터 원래의 파형과 비교하여 2채널의 음원신호를 깨끗하게 분리할 수 있음을 명확히 하였다. 또한 목표 신호 대 간섭 에너지비율을 사용하여 비교한 실험 결과로부터 본 논문에서 제안한 알고리즘의 음원분리 성능이 기존의 알고리즘에 비하여 성능이 향상되었다는 것을 알 수 있었다.

This paper proposes a blind speech source separation algorithm using a microphone to separate only the target speech source signal in an environment in which various speech source signals are mixed. The proposed algorithm is a model of frequency domain representation based on independent component analysis method. Accordingly, for the purpose of verifying the validity of independent component analysis in the frequency domain for two speech sources, the proposed algorithm is executed by changing the type of speech sources to perform speech sources separation to verify the improvement effect. It was clarified from the experimental results by the waveform of this experiment that the two-channel speech source signals can be clearly separated compared to the original waveform. In addition, in this experiments, the proposed algorithm improves the speech source separation performance compared to the existing algorithms, from the experimental results using the target signal to interference energy ratio.

키워드

참고문헌

  1. K. Nakadai, H. Nakajima, G. Ince, and Y. Hasegawa, "Sound source separation and automatic speech recognition for moving sources," IEEE/RSJ Int. Conference on Intelligent Robots and Systems, Taipei, Taiwan, Oct., 2010, pp. 976-981.
  2. T. Kim, H. T. Attias, S. Y. Lee, and T. W. Lee, "Blind Source Separation Exploiting Higher-Order Frequency Dependencies," IEEE Transactions on Audio, Speech, and Language Processing, vol. 15, no. 1, Jan. 2007, pp. 70-79. https://doi.org/10.1109/TASL.2006.872618
  3. H. T. Kim, "Vocal Separation in Music Using SVM and Selective Frequency Subtraction," J. of Korea Institute of Electronic Communication Sciences, vol. 10, no. 1, 2015, pp. 1-6. https://doi.org/10.13067/JKIECS.2015.10.1.1
  4. C. B. Lee, "Evaluation of a signal segregation by FDBM," J. of the Korea Institute of Electronic Communication Sciences, vol. 8, no. 12, 2013, pp. 1793-1802. https://doi.org/10.13067/JKIECS.2013.8.11.1793
  5. J. S. Choi, "Mixed Noise Cancellation by Independent Vector Analysis and Frequency Band Beamforming Algorithm in 4-channel Environments," J. of the Korea Institute of Electronic Communication Sciences, vol. 14, no. 5, 2019, pp. 811-816.
  6. F. Asano, S. Ikeda, M. Ogawa, H. Asoh, and N. Kitawaki, "Combined approach of array processing and independent component analysis for blind separation of acoustic signals," IEEE Trans. on Speech and Audio Processing, vol. 11, no. 3, May 2003, pp. 204-215. https://doi.org/10.1109/TSA.2003.809191
  7. H. Sawada, S. Araki, and S. Makino, "Measuring Dependence of Bin-Wise Separated Signals for Permutation Alignment in Frequency-Domain BSS," IEEE International Symposium on Circuits and Systems, New Orleans, LA, USA, May 2007, pp. 3247-3250.
  8. Z. Chu and K. S. Bae, "Post-processing of IVA-based 2-channel blind source separation for solving frequency bin permutation problem," Phonetics and Speech Sciences, vol. 5, no. 4, Dec. 2013, pp. 211-216. https://doi.org/10.13064/KSSS.2013.5.4.211
  9. X. Wang, X. Quan, and K. S. Bae, "Microphone Array Based Speech Enhancement Using Independent Vector Analysis," Phonetics and Speech Sciences, vol. 4, no. 4, Dec. 2012, pp. 87-92. https://doi.org/10.13064/KSSS.2012.4.4.087
  10. J. S. Choi, "A Blind Source Separation Method Based on Independent Vector Analysis for Separation of Speech Signal and Noise Signal," The Journal of Korean Institute of Information Technology, vol. 16, no. 10, Oct. 2018, pp. 69-74. https://doi.org/10.14801/jkiit.2018.16.10.69
  11. E. Bingham and A. Hyvarinen, "A fast fixed-point algorithm for independent component analysis for complex valued signals," International Journal of Neural Systems, vol. 10, no. 1, Feb. 2000, pp. 1-8. https://doi.org/10.1142/S0129065700000028
  12. T. Nishikawa, H. Saruwatari, and K. Shikano, "Comparison of time-domain ICA, frequency-domain ICA and multistage ICA for blind source separation," 2002 11th European Signal Processing Conference, vol. II, Sept. 2002, pp. 15-18.
  13. F. Nesta, P. Svaizer, and M. Omologo, "Convolutive BSS of short mixtures by ICA recursively regularized across frequencies," IEEE Trans. on Audio, Speech, and Language Processing, vol. 19, no. 3, Mar. 2011, pp. 624-639. https://doi.org/10.1109/TASL.2010.2053027
  14. X. Quan and K. S. Bae, "Improvement of convergence speed in FDICA algorithm with weighted inner product constraint of unmixing matrix," Phonetics and speech sciences, vol. 7, no. 4, 2015, pp. 17-25. https://doi.org/10.13064/KSSS.2015.7.4.017
  15. H. W. Lee, "Acoustic Echo Cancellation Based on Convolutive Blind Signal Separation Method," J. of the Korea Institute of Electronic Communication Sciences, vol. 13, no. 5, Oct. 2018, pp. 979-986. https://doi.org/10.13067/JKIECS.2018.13.5.979