Abstract
This paper presents a new algorithm for noise reduction in noisy speech recognition in autocorrelation domain. The autocorrelation domain is an appropriate domain for speech feature extraction due to its pole preserving and noise separation features. Therefore, we have investigated this domain for robust speech recognition.
In our proposed algorithm we have tried to suppress the effect of noise before using this domain for feature extraction. This suppression is carried out by noise autocorrelation sequence estimation from the first few frames in each utterance and subtracting it from the autocorrelation sequence of noisy signal. We tested our method on the Aurora 2 noisy isolated-word task and found its performance superior to that of other autocorrelation-based methods applied to this task.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Mansour, D., Juang, B.-H.: The Short-time Modified Coherence Representation and Noisy Speech Recognition. IEEE Trans. on Acoustics, Speech and Signal Processing 37(6), 795–804 (1989)
Hernando, J., Nadeu, C.: Linear Prediction of the One-sided Autocorrelation Sequence for Noisy Speech Recognition. IEEE Trans. Speech and Audio Processing 5(1), 80–84 (1997)
You, K.-H., Wang, H.-C.: Robust Features for Noisy Speech Recognition Based on Temporal Trajectory Filtering of Short-time Autocorrelation Sequences. Speech Communication 28, 13–24 (1999)
Shannon, B.-J., Paliwal, K.-K.: MFCC Computation from Magnitude Spectrum of Higher lag Autocorrelation Coefficients for Robust Speech Recognition. In: Proc. ICSLP, pp. 129–132 (2004)
Farahani, G., Ahadi, S.M.: Robust Features for Noisy Speech Recognition Based on Filtering and Spectral Peaks in Autocorrelation Domain. In: Proc. EUSIPCO, Antalya, Turkey (2005)
Ikbal, S., Misra, H., Bourlard, H.: Phase autocorrelation (PAC) derived robust speech features. In: Proc. ICASSP, Hong Kong, pp. II-133–II-136 (2003)
McGinn, D.-P., Johnson, D.-H.: Estimation of all-pole model parameters from noise-corrupted sequence. IEEE Trans. on Acoustics, Speech and Signal Processing 37(3), 433–436 (1989)
Chen, J., Paliwal, K.-K., Nakamura, S.: Cepstrum derived from differentiated power spectrum for robust speech recognition. Speech Communication 41, 469–484 (2003)
The hidden Markov model toolkit, available from, http://htk.eng.cam.ac.uk
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Farahani, G., Ahadi, S.M., Homayounpour, M.M. (2006). Robust Feature Extraction of Speech Via Noise Reduction in Autocorrelation Domain. In: Gunsel, B., Jain, A.K., Tekalp, A.M., Sankur, B. (eds) Multimedia Content Representation, Classification and Security. MRCS 2006. Lecture Notes in Computer Science, vol 4105. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11848035_62
Download citation
DOI: https://doi.org/10.1007/11848035_62
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-39392-4
Online ISBN: 978-3-540-39393-1
eBook Packages: Computer ScienceComputer Science (R0)