Modulation spectrum factorization for robust speech recognition

Wen Yi Chu, Jeih Weih Hung, Berlin Chen

研究成果: 會議貢獻類型

7 引文 斯高帕斯(Scopus)

摘要

This paper presents a novel approach to improving the noise robustness of speech features built on top of nonnegative matrix factorization (NMF). To do this, we employ NMF to extract a common set of basis spectral vectors that cover the intrinsic temporal structure inherent in the modulation spectra of clean training speech features. The new modulation spectra of the speech features, constructed by mapping the original modulation spectra into the space spanned by these basis vectors, are demonstrated with good noise-robust capabilities. All experiments were conducted using the Aurora-2 database and task. The results show that the proposed NMF-based approach, together with mean and variance normalization (MVN), can provide average error reduction rates of over 65% and 12% relative as compared with the baseline MFCC system and that using the MVN method alone, respectively.

原文英語
頁面1-6
頁數6
出版狀態已發佈 - 2011 十二月 1
事件Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011, APSIPA ASC 2011 - Xi'an, 中国
持續時間: 2011 十月 182011 十月 21

其他

其他Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011, APSIPA ASC 2011
國家中国
城市Xi'an
期間11/10/1811/10/21

    指紋

ASJC Scopus subject areas

  • Information Systems
  • Signal Processing

引用此

Chu, W. Y., Hung, J. W., & Chen, B. (2011). Modulation spectrum factorization for robust speech recognition. 1-6. 論文發表於 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011, APSIPA ASC 2011, Xi'an, 中国.