A New Feature Extraction Method for Bone-Conducted Life Sounds Based on F-Ratio

Abstract:

Bone-conducted life sounds are useful for monitoring human healthy situation. Although a number of feature extraction methods were proposed for air-conducted speech, they may not meet the requirements of the recognition task for bone-conducted life sounds since there is a large difference between air-conducted speech and bone-conducted life sounds. In order to obtain features that can characterize bone-conducted signals, in this study, we first analyze the property of bone-conducted life sounds itself and compare each kind of life sounds in the frequency region. Then we adopt the methods of F-ratio and improved F-ratio separately to measure the dependences between frequency components and characteristics of life sounds. According to the result of analysis, we design a new adaptive frequency filter to extract the desired discriminative feature. The new feature is combined with the Hidden Markov Model and applied to classify different kinds of bone-conducted life sounds. The experimental results show that the error rate using the proposed feature based on State mean F-ratio is reduced by 7.2% compared with the MFCC feature.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 1598-1604
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.1598
• Online since: September 2013
• Authors: Ye Teng An, Hong Cui Wang, Song Gun Hyon, Sai Chen, Jian Wu Dang
• Keywords: Bone-Conducted, Discriminative Feature, Feature Extraction, F-Ratio, Life Sounds

References

[1] Wang, J., Zhang, X. Study and simulation of the acoustic features in speaker recognition system. Journal of System Simulation 15(9), 1276-1278. (2003).

[2] Rui, X., Yu, Y. A combined feature extraction method for Speaker identification under noisy conditions. Signal Processing(In Chinese) 22(5), 673-677. (2006).

[3] Hsieh, C.T., Lai, E., Wang, Y.C. Robust speech features based on wavelet transform with application to speaker identification. IEE Proceedings: Vision, Image and Signal Processing 149(2), 108-144. (2002).

DOI: 10.1049/ip-vis:20020121

[4] Zhang, L., Zheng, B., Yang, Z. Robust feature based on speech harmonic structure for speaker identification. Journal of Electronics & Information Technology(In Chinese) 28(10), 1786-1789. (2006).

[5] Hayakawa, S., Itakura, F. Text-dependent speaker recognition using the information in the higher frequency band. In: Proc. ICASSP1994, pp. I-137-I-140. (1995).

[6] Wolf, J.J. Efficient acoustic parameters for speaker recognition. J. Acoust. Soc. Am. 51, 2044–2056. (1972).

[7] Xugang Lu, Jianwu Dang, An investigation of dependencies between frequency components and speaker characteristics for text-independent speaker identification,. Speech Communication. 2008, 50: 312-322.

DOI: 10.1016/j.specom.2007.10.005

[8] Songgun Hyon, Hongcui Wang, Chen Zhao, Jianwu Dang. A method of speaker identification based on phoneme mean F-ratio contribution, In Proc. INTERSPEECH, Portland, U.S. A, (2012).

DOI: 10.21437/interspeech.2012-349

[9] Tetsuya Shimamura, Jun'ichiro Mamiya and Toshiki Tamiya. Improving Bone-Conducted Speech Quality via Neural Network, Signal Processing and Information Technology, (2006).

DOI: 10.1109/isspit.2006.270876

[10] L. R. Rabiner, A tutorial on hidden Markov models and selected applications in speech recognition, Proc. IEEE, vol. 77, p.257–286, Feb. (1989).

DOI: 10.1109/5.18626