Improved perceptually non-uniform spectral compression for 
robust speech recognition

doi:10.1016/S1005-8885(13)60080-1

中国邮电高校学报(英文) ›› 2013, Vol. 20 ›› Issue (4): 122-126.doi: 10.1016/S1005-8885(13)60080-1

Improved perceptually non-uniform spectral compression for robust speech recognition

张毅¹,何春江¹,罗元²,陈凯¹,邢武超¹

Engineering Research and Development Centre of Information Accessibility, Chongqing University of Posts and Telecommunications, Chongqing 40065, China

收稿日期:2013-01-15 修回日期:2013-04-16 出版日期:2013-08-31 发布日期:2013-08-30
通讯作者: 何春江 E-mail:hcj460@163.com
基金资助:
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).

Improved perceptually non-uniform spectral compression for robust speech recognition

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 E-mail:hcj460@163.com
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).

摘要/Abstract

摘要： 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.

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

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

中图分类号:

TP311

ZHANG Yi, HE Chun-jiang, LUO Yuan, CHEN Kai, XING Wu-chao. Improved perceptually non-uniform spectral compression for robust speech recognition [J]. Acta Metallurgica Sinica(English letters), 2013, 20(4): 122-126.

参考文献

1. Zou D Y, Li L. Research on robust speech feature parameter extraction in colored noise environment. Computer Simulation, 2011, 28(5): 395-398 (in Chinese)

2. Hung J W, Tu W H, Lai C C. Improved modulation spectrum enhancement methods for robust speech recognition. Signal Processing, 2012, 92(11): 2791-2814

3. Lü Z. Study of speech recognition algorithm under noise environment. Hefei, China: Anhui University, 2011: 57-60 (in Chinese)

4. Shajith I, Hemant M, Hynek H, et al. Phase auto correlation (PAC) features noise robust speech recognition. Speech Communication, 2012, 54(7): 867-880

5. Zhao H, Zhao K, Liu H, et al. Improved MFCC feature extraction combining symmetric ICA algorithm for robust speech recognition. Journal of Multimedia, 2012, 7(1): 74-81

6. Lü Y, Wu Z Y. Robust speech recognition based on vector taylor series. Journal of Tianjin University, 2011, 44(3): 261-265 (in Chinese)

7. Lei J J. A study of several problems on noise robust speech recognition. Beijing: Beijing University of Posts and Telecommunications, 2007: 12-26 (in Chinese)

8. Patrice A, Philip L. Root cepstral analysis: a unified view. application to speech processing in car noise environments. Speech Communication, 1993, 12(3): 277-288

9. Cai L H, Huang D Z, Cai R. The foundation and application of modern speech technology. Beijing, China: Tsinghua University Press, 2003: 233-237 (in Chinese)

10. Luo Y, Huang H, Zhang Y, et al. A novel endpoint detection method and its application on human-computer interaction of intelligent wheelchair. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2011, 23(2): 86-90 (in Chinese)

11. Ning G X. Research on novel robust speech recognition. Guangzhou, China: South China University of Technology, 2006: 31-34 (in Chinese)

12. Ding P. Noise robust technologies in speech recognition. Beijing, China: Tsinghua University, 2003: 32-40 (in Chinese)

13. Long Q. Research on speech recognition in noisy environment. Hefei, China: University of Science and Technology of China, 2007: 81-85 (in Chinese)

14. Zhang L. The research of robust speech recognition in noise environment based on MFCC. Changsha, China: Hunan University, 2009: 21-26 (in Chinese)

15. Chu K K. Feature extraction based on perceptual non-uniform spectral compression for noisy speech recognition. Hong Kong, China: City University of Hong Kong, 2005: 82-91

16. Chu K K, Leung S H, Yip C S. Perceptually non-uniform spectral compression for noisy speech recognition. Proceedings of the 28th International Conference on Acoustics, Speech, and Signal Processing (ICASSP’03): Vol 1, Apr 6-10, 2003, Hong Kong, China. Piscataway, NJ, USA: IEEE: 404-407

17. Strvens S S. The direct estimation of sensory magnitudes: Loudness. The American Journal of Psychology, 1956, 69(1): 1-25

18. Hartmann W M. Signals, sounds, and sensation. Berlin, Germany: Springer-Verlag, 1997: 238-256

Improved perceptually non-uniform spectral compression for robust speech recognition

Improved perceptually non-uniform spectral compression for robust speech recognition

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价