期刊文献+

Speech wideband extension based on Gaussian mixture model 被引量:4

Speech wideband extension based on Gaussian mixture model
原文传递
导出
摘要 To decrease the spectral distortion of highband envelope, the function of spectral distortion and mutual information between feature vector and highband envelope was studied, and an extended Gaussian Mixture Model (GMM) bandwidth extension algorithm was proposed based on the research. The feature parameters which have larger mutual information with highband envelope were selected to constitute the feature vector, and the GMM was adopted to compute the joint probability density of the feature vector and highband envelope. Then the highband envelope was estimated via the posterior probabilities computed from the model parameters estimated by Expectation-Maximization (EM) algorithm. The experimental results show that the spectral distortion is lower than the algorithm, such as the traditional algorithm based on GMM, by 0.3 dB and the number of frames with spectral distortion over 10 dB sharply reduced over 50%. To decrease the spectral distortion of highband envelope, the function of spectral distortion and mutual information between feature vector and highband envelope was studied, and an extended Gaussian Mixture Model (GMM) bandwidth extension algorithm was proposed based on the research. The feature parameters which have larger mutual information with highband envelope were selected to constitute the feature vector, and the GMM was adopted to compute the joint probability density of the feature vector and highband envelope. Then the highband envelope was estimated via the posterior probabilities computed from the model parameters estimated by Expectation-Maximization (EM) algorithm. The experimental results show that the spectral distortion is lower than the algorithm, such as the traditional algorithm based on GMM, by 0.3 dB and the number of frames with spectral distortion over 10 dB sharply reduced over 50%.
出处 《Chinese Journal of Acoustics》 2009年第4期362-377,共16页 声学学报(英文版)
  • 相关文献

参考文献2

二级参考文献15

  • 1崔锦泰 程正兴(译).小波分析导论[M].西安交通大学出版社,1994..
  • 2程正兴(译),小波分析导论,1995年
  • 3Sang Enfang,Proc. of HICEC’92,1992年,715页
  • 4Yang X,IEEE Trans Information Theory,1992年,38卷,2期,824页
  • 5陈永彬,语言信号处理,1990年
  • 6Cheng Y M, O'Shaugnessy D, Mermelstein P. Statistical recovery of wideband speech from narrowband speech[J]. IEEE Transaction Speech Audio Process, 1994(2): 544-548.
  • 7Park K Y, Kim H S. Narrowband to wideband conversion of speech using GMM based transformation[Z]. International Conference on Acoustic Speech Signal Process, Istanbul, 2000.
  • 8Jax P, Vary P. Wideband extension of telephone speech using a hidden markov model[A]. IEEE Workshop on Speech Coding[C], Delavan: IEEE,2000.
  • 9Yoshida Y, Abe M. An algorithm to reconstruct wideband speech from narrow band speech based on codebook mapping[Z]. IEEE International Conference on Spoken Language Processing, Yokohama, 1994.
  • 10Enbom N, Kleijn W B. Bandwidth expansion of speech based on vector quantization of the mel frequency cepstral coefficients[A]. IEEE Workshop on Speech Coding[C], Porvoo, Finland: IEEE,1999.

共引文献9

同被引文献14

引证文献4

二级引证文献8

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部