Acta Metallurgica Sinica(English letters) ›› 2013, Vol. 20 ›› Issue (4): 122-126.doi: 10.1016/S1005-8885(13)60080-1

• Others • Previous Articles     Next Articles

Improved perceptually non-uniform spectral compression for robust speech recognition


  1. Engineering Research and Development Centre of Information Accessibility, Chongqing University of Posts and Telecommunications, Chongqing 40065, China
  • Received:2013-01-15 Revised:2013-04-16 Online:2013-08-31 Published:2013-08-30
  • Contact: Chun-Jiang HE
  • Supported by:
    This work was supported by the International Science and Technology Cooperation Program of China (2010DFA12160), the National Natural Science Foundation of China (51075420), the National Natural Science Foundation of China (60905066), and the Science & Technology Research Project of Chongqing CSTC (2010AA2055).

Abstract: According to the decline of recognition rate of speech recognition system in the noise environments, an improved perceptually non-uniform spectral compression feature extraction algorithm is put forward in this paper. This method can realize an effective compression of the speech signals and make the training and recognition environments more matching, so the recognition rate can be improved in the noise environments. By experimenting on the intelligent wheelchair platform, the result shows that the algorithm can effectively enhance the robustness of speech recognition, and ensure the recognition rate in the noise environments.

Key words: robust speech recognition, improved perceptually non-uniform spectral compression, intelligent wheelchair, mel-frequency cepstrum coefficients

CLC Number: