Abstract
In this paper, a new speech enhancement based on the nonlinear H ∞ filtering and neural predictive HMM (NPHMM) is presented. In H ∞ filtering, no a priorknowledge of the noise source statistics is required. Speech is modeled as the output of a neural predictive HMM combining MLP neural network and HMM. The proposed enhancement method consists of multiple nonlinear H ∞ filters with parameter of the NPHMM. The switching between the nonlinear H ∞ filters is governed by a finite state Markov chain according to the transition probabilities. An approximate improvement of 0.4-1.8dB in output SNR is achieved at various input SNRs compared with conventional Kalman method.
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
Ephraim, Y.: A Bayesian estimation approach for speech enhancement using HMM. IEEE Trans. Signal Proc. 41, 725–735 (1993)
Lee, K.Y., et al.: Speech enhancement based on neural predictive hidden Markov model. Signal Processing 65, 373–381 (1998)
Shen, X., Deng, L.: A dynamic system approach to speech enhancement using the H∞ filtering algorithm. IEEE Trans. Speech and Audio Proc. 7, 391–399 (1999)
Banavar, R.N., Speyer, J.L.: A linear quadratic game theory approach to estimation and smoothing. In: Proc. IEEE ACC, pp. 2818–2822 (1991)
Nishiyama, K., Suzuki, K.: H∞ -learning of layered neural networks. IEEE Trans. Neural Networks 12, 1265–1277 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lee, K.Y., Rheem, J.Y. (2005). Robust Speech Enhancement Based on NPHMM Under Unknown Noise. In: Chollet, G., Esposito, A., Faundez-Zanuy, M., Marinaro, M. (eds) Nonlinear Speech Modeling and Applications. NN 2004. Lecture Notes in Computer Science(), vol 3445. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11520153_29
Download citation
DOI: https://doi.org/10.1007/11520153_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-27441-4
Online ISBN: 978-3-540-31886-6
eBook Packages: Computer ScienceComputer Science (R0)