DOI QR코드

DOI QR Code

SMV코덱의 음성/음악 분류 성능 향상을 위한 최적화된 가중치를 적용한 입력벡터 기반의 SVM 구현

Analysis and Implementation of Speech/Music Classification for 3GPP2 SMV Codec Employing SVM Based on Discriminative Weight Training

  • 김상균 (인하대학교 전자공학부) ;
  • 장준혁 (인하대학교 전자공학부) ;
  • 조기호 (서울대학교 전기컴퓨터공학부) ;
  • 김남수 (서울대학교 전기컴퓨터공학부)
  • 발행 : 2009.07.31

초록

본 논문에서는 변별적 가중치 학습 (discriminative weight training) 기반의 최적화된 가중치를 가지는 입력벡터를 구성하여 support vector machine (SVM)을 이용한 기존의 3GPP2 selectable mode vocoder (SMV)코덱의 음성/음악 분류 성능을 향상 시키는 방법을 제안한다. 구체적으로, 최소 분류 오차 minimum classification error (MCE) 방법을 도입하여, 최적화된 가중치를 각각의 특징벡터별로 부가한 SVM을 적용하여 기존의 가중치를 고려하지 않은 SVM 기반의 알고리즘과 비교하였으며, 우수한 음성/음악 분류 성능을 보였다.

In this paper, we apply a discriminative weight training to a support vector machine (SVM) based speech/music classification for the selectable mode vocoder (SMV) of 3GPP2. In our approach, the speech/music decision rule is expressed as the SVM discriminant function by incorporating optimally weighted features of the SMV based on a minimum classification error (MCE) method which is different from the previous work in that different weights are assigned to each the feature of SMV. The performance of the proposed approach is evaluated under various conditions and yields better results compared with the conventional scheme in the SVM.

키워드

참고문헌

  1. Y. Gao, E. Shlomot, A. Benyassine, J. Thyssen, Huan-yu Su and C. Murgia, "The SMV Algorithm Selected by TIA and 3GPP2 for CDMA Apphcations," Proc. IEEE International Con-ference on Acoustics, Speech and Siginal Processing, vol. 2, pp. 709-712, May 2001 https://doi.org/10.1109/ICASSP.2001.941013
  2. 3GPP2 Spec.. "Source-controlled variable-rate multimedia wideband speech codec (VMR-WB), service option 62 and 63 for spread spectrum systems," 3GPP2-C.S0052-A, v.1.0, Apr. 2005
  3. J. Saunders, "Real-time discrimination of broadcast speech/music," Proc. IEEE International Conference on Acoustics, Speech, and Processing, vol. 2, pp. 993-996, May 1996 https://doi.org/10.1109/ICASSP.1996.543290
  4. W. Q. Wang, W. Gao arid D. W. Ying, "A fast and robust speech/music Discrimination Approach," Proc. International Conference on Information, Communications and Singnal Pro-cessing, vol. 3, pp. 1325-1329, Dec. 2003 https://doi.org/10.1109/ICICS.2003.1292679
  5. 김상균, 장준혁, "SMV코덱의 음성/음악 분류 성능 향상을 위한 Support Vector Machine의 적용" , 전자공학회논문지, 제45권, 제6호, pp.142-147, 2008
  6. 3GPP2 Spec., "Selectable Mode Vocoder (SW) Service Option forr Wideband Spread Spectrum Communication Systems," 3GPP2-C.S0030-0, v3.0, Jan. 2004
  7. V. N. Vapnik, "An overview of statistical learning theory," IEEE Trans. Neural Networks, vol. 10, no. 5, pp. 988-999, 1999 https://doi.org/10.1109/72.788640
  8. N. Cristianini and J. Shawe-Taylor, An introduction to su-pport vector machines and other kernel-based learning methods. Cambridge Univ. Press, 2000
  9. B.-H. Juang. W. Chou and C.-H. Lee, "Minimum classifi-cation error rate methods for speech recognition," IEEE Trans. Speech Audio Processing, vol. 5, no. 3, pp. 257-265, 1997 https://doi.org/10.1109/89.568732
  10. S.-l. Kang, Q.-H. Jo and J.-H, Chang, "Discriminative weight training for a statistical model-based voice activity detection," IEEE SignaI Processing Letter, vol. 15, pp. 170-173, 2008 https://doi.org/10.1109/LSP.2007.913595