Combining Speech Enhancement and Cepstral Mean Normalization for LPC Cepstral Coefficients

Abstract:

A mismatch between the training and testing in noisy circumstance often causes a drastic decrease in the performance of speech recognition system. The robust feature coefficients might suppress this sensitivity of mismatch during the recognition stage. In this paper, we investigate the noise robustness of LPC Cepstral Coefficients (LPCC) by using speech enhancement with feature post-processing. At front-end, speech enhancement in the wavelet domain is used to remove noise components from noisy signals. This enhanced processing adopts the combination of discrete wavelet transform (DWT), wavelet packet decomposition (WPD), multi-thresholds processing etc to obtain the estimated speech. The feature post-processing employs cepstral mean normalization (CMN) to compensate the signal distortion and residual noise of enhanced signals in the cepstral domain. The performance of digit speech recognition systems is evaluated under noisy environments based on NOISEX-92 database. The experimental results show that the presented method exhibits performance improvements in the adverse noise environment compared with the previous features.

• Periodical: Key Engineering Materials (Volumes 474-476)
• Pages: 349-354
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.474-476.349
• Online since: April 2011
• Authors: Jie Yang
• Keywords: Cepstral Mean Normalization, LPCC, Speech Enhancement, Speech Recognition

References

[1] L.H. Chang, L.S. Young: Noise-Robust Speech Recognition Using Top-Down Selective Attention With an HMM Classifier. IEEE Signal Processing Letters Vol. 14 (2007), pp.489-491.

DOI: 10.1109/lsp.2006.891326

[2] C. Xiaodong, G. Yifan: A Study of Variable- Parameter Gaussian Mixture Hidden Markov Modeling for Noisy Speech Recognition. IEEE Trans. on Audio, Speech, and Language Processing Vol. 15 (2007), pp.1366-1376.

DOI: 10.1109/tasl.2006.889791

[3] J.W. Picone: Signal modeling techniques in speech recognition. Proc. IEEE, Vol. 81 (1993), pp.1215-1247.

DOI: 10.1109/5.237532

[4] H. Hermansky: Perceptual linear predictive (PLP) analysis of speech. J. Acoust. Soc. Am. Vol. 87 (1990), pp.1738-1752.

[5] H. Hermansky, N. Morgan: RASTA Processing of Speech. IEEE Trans. on Speech and Audio Processing Vol. 2 (1994), pp.578-589.

DOI: 10.1109/89.326616

[6] W. Zhenli, Z. Xiongwei, Z. Xiang: A new wavelet domain speech enhancement method. Signal Processing Vol. 22(2006), pp.325-328. (in Chinese).

[7] F.H. Liu, A. Acero, and R. Stern: Efficient Joint Compensation of Speech For the Effects of Additive Noise and Linear Filtering. IEEE International Conference on Acoustics, Speech, and Signal Processing Vol. 1 (1992), pp.257-260.

DOI: 10.1109/icassp.1992.225923

[8] O. Viildu, D. Bye, K. Iaurila: A recursive feature vector normalization approach for robust speech recognition in noise [A]. Proceedings'ICASSP'98 [C]. Seattle, WA, USA: IEEE Acoustics, Speech and Signal Processing Society, 1998, pp.733-736.

DOI: 10.1109/icassp.1998.675369