A Speech Feature Extraction Algorithm Based on the Nonlinear Hopf Cochlea Model

Abstract:

The performance of speech recognition system is well or not is closely related to the characteristic parameters. For emulating human auditory system, a new method of speech feature extraction based on Hopf filter banks is presented. We modeled the extraction process of the MFCC, and used Hopf filter banks instead of the triangular filter banks. Then, we according the characteristics of Basilar Membranes in the cochlea to adjust the center frequency and bandwidth of the filter. The test speech goes through the Hopf filter banks, multi-dimensional eigenvectors will be obtained. After that, by Discrete Cosine Transformation, we will get the Hopf cepstral coefficients of the speech. Comparing with traditional feature MFCC, the speech recognition systems with Hopf characteristic parameters have better recognition rate and robustness characteristics in low Signal Noise Ratio (SNR) environment.

• Periodical: Applied Mechanics and Materials (Volumes 416-417)
• Pages: 1176-1180
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.416-417.1176
• Online since: September 2013
• Authors: Jian Guo Xing, Min Xu, Ji Xiang Zhu
• Keywords: Hopf Filter Banks, Index Terms Speech Feature Extraction, Recognition Rate, SNR

References

[1] T. J. Hamilton, J. Tapson, C. Jin, and A. van Schaik (2007) A 2-D Cochlea with Hopf Oscillators, in IEEE Biomedical Circuits and Systems Conference Montreal, Canada 14(5): 189-182.

DOI: 10.1109/biocas.2007.4463316

[2] T. J. Hamilton, C. Jin, and A. van Schaik (2008) Analogue VLSI Implementations of Two Dimensional, Nonlinear, Active Cochlea Models, Biomedical Circuits and Systems Conference. IEEE 15(4): 153-156.

DOI: 10.1109/biocas.2008.4696897

[3] T. J. Hamilton, C. Jin, A. van Schaik, and J. Tapson (2008) An Active 2-D Silicon Cochlea. Biomedical Circuits And Systems, IEEE Transactions on 2(11): 30-43.

DOI: 10.1109/tbcas.2008.921602

[4] Glasberg B. R, Moore B.C. J (1900) Derivation of auditory filter shapes from notched-noise data Hearing Research 47(12): 103-108.

DOI: 10.1016/0378-5955(90)90170-t

[5] S laney M (1993) An efficient implementation of the Patterson-Hold sworth auditory filter bank Apple Computer Inc 15(7): 25-32.

[6] T.J. Hamilton, J . Tapson, M. Rapson, C. Jin, and A. van Schaik (2010) Understanding the mathematics of hearing using electronic circuits. Anziam J 51(4): 300-315.

DOI: 10.21914/anziamj.v51i0.2543