Correction and Interpolation of Noise Corrupted Voice Using Markov Chain Detection Technique

In this paper, we present a technique for voice correction and interpolation using Markov chain detection. In a communication link, the voice is impaired by spectral distortions generated by the noise and fading. Our purpose is to restore voice samples as close as possible to that of the original one of the speaker using a Markov chain detection technique, which compensates for the spectral distortions. Assuming that, voice maintains high order memory, this has been exploited for restoration of distorted voice samples. By formulating the problem of missing or distorted voice samples by Markov chain Detection Technique, we show that it is possible to correct and interpolate the degraded signal close to that of original one. The performance of the restored signal is evaluated by few simulations using spectrogram analysis.

[1]  Simon J. Godsill,et al.  Interpolation of missing data values for audio signal restoration using a Gabor regression model , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[2]  Svetha Venkatesh,et al.  Topic transition detection using hierarchical hidden Markov and semi-Markov models , 2005, MULTIMEDIA '05.

[3]  Saeed Vaseghi,et al.  Noise compensation methods for hidden Markov model speech recognition in adverse environments , 1997, IEEE Trans. Speech Audio Process..

[4]  R.A. Shafik,et al.  On the Lempel-Ziv-Welch coded speech transmission and markov chain detection , 2006, 2006 International Conference on Emerging Technologies.

[5]  J. Picone,et al.  Continuous speech recognition using hidden Markov models , 1990, IEEE ASSP Magazine.

[6]  Keiichi Tokuda,et al.  Hidden Markov models based on multi-space probability distribution for pitch pattern modeling , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[7]  Michael A. Soderstrand,et al.  Objective measurement of voice quality for cellular mobile phones , 1998, 1998 Midwest Symposium on Circuits and Systems (Cat. No. 98CB36268).

[8]  Gerardo Rubino,et al.  Performance evaluation of real-time speech through a packet network: a random neural networks-based approach , 2004, Perform. Evaluation.

[9]  Ho-Chung Lui,et al.  A multilayer perceptron postprocessor to hidden Markov modeling for speech recognition , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[10]  Paulo A. A. Esquef,et al.  Interpolation of Long Gaps in Audio Signals Using Line Spectrum Pair Polynomials , 2004 .