SIMULATION OF MODIFIED HYBRID NOISE REDUCTION ALGORITHM TO ENHANCE THE SPEECH QUALITY
Abstract
Speech is the most essential method of correspondence of humankind. Cell telephony, portable hearing assistants and, hands free are specific provisions in this respect. The performance of these communication devices could be affected because of distortions which might augment them. There are two essential sorts of distortions that might be recognized, specifically: convolutive and additive noises. These mutilations contaminate the clean speech and make it unsatisfactory to human audiences' i.e. perceptual value and intelligibility of speech signal diminishes. The objective of speech upgrade systems is to enhance the quality and understandability of speech to make it more satisfactory to audiences. This paper recommends a modified hybrid approach for single channel devices to process the noisy signals considering only the effect of background noises. It is a mixture of pre-processing relative spectral amplitude (RASTA) filter, which is approximated by a straightforward 4th order band-pass filter, and conventional minimum mean square error short time spectral amplitude (MMSE-STSA85) estimator. To analyze the performance of the algorithm an objective parameter called Perceptual estimation of speech quality (PESQ) is measured. The results show that the modified algorithm performs well to remove the background noises. SIMULINK implementation is also performed and its profile report has been generated to observe the execution time.References
B. Gold and N. Morgan, Speech and Audio
Signal Processing, John Wiley & Sons, Inc.
(2000) Vii, 3, 4, 15.
C. H. You, S. N. Koh and S. Rahardja,
Adaptive β-order MMSE Speech
Enhancement Application for Mobile
Communication in a Car Environment, in
Proc. 4th Int. Conf. on Information,
Communications and Signal Processing, 4
th
Pacific Rim Conf. on Multimedia (2003) pp.
–1632.
T. Yamada, M. Kumakura and N. Kitawaki,
IEEE Trans. Audio, Speech, Language
Processing 14, Nov. (2006) 2006.
J. Benesty, S. Makino and J. Cheng, Speech
Enhancement, Springer Series of Signals
and Communication Technology (2005).
F. Toledo, P. C. Loizou and A. Lobo,
Subspace and Envelope Subtraction
Algorithms for Noise Reduction in Cochlear
Implants, IEEE Annual International
Conference of Engineering Medicine and
Biology Society (2003) pp. 213-216.
S. F. Boll, IEEE Trans. on Acoustic, Speech
and Signal Processing, ASSP-27, April
(1979) 113.
Y. Ephraim and D. Malah, IEEE Trans. on
Acoustics, Speech and Signal Processing,
ASSP-32 (1984) 1109.
Y. Ephraim and D. Malah, IEEE Trans. on
Acoustics, Speech and Signal Processing,
ASSP-33 (1985) 443.
C. Breithaupt and R. Martin, MMSE
Estimation of Magnitude-squared dft
Coefficients with Super Gaussian Priors in
Proc. IEEE Int. Conf. Acoust, Speech, Signal
Processing (2003) pp. 848-851.
H. Hermanskey and N. Morgan, IEEE Trans.
on Acoustics, Speech and Signal Processing
(1994) 578.
S.K. Shah, J.H. Shah and N.N.Parmar,
Evaluation of RASTA Approach with Modified
Parameters for Speech Enhancement in
Communication Systems, Proc. IEEE
Symposium on Computers and Informatics
(March 2011) pp. 159-162.
J. Shah and S. Shah, International Journal of
Computer Science 9, Issue 4, No 2 (2012)
S.K. Shah and J.H. Shah, Real Time and
Embedded Implementation of Hybrid
Algorithm for Speech Enhancement,
Information and Communication Technologies (WICT), Mumbai, India (2011) pp. 341-
D. Burshtein and S. Gannot, IEEE Trans.
Speech, Audio Process 10 (2002) 341.
The NOIZEUS database (2009) available at:
http://www.utdallas.edu/~loizou/speech/noize
A. Hu and P. Loizou, Subjective
Comparisons of Speech Enhancement
Algorithms, Proc. IEEE International
conference on Acoustics, Speech and Signal
Processing (May 2006).