|
Abstract
|
Article Information:
Noise Minimization from Speech Signals using RLS Algorithm with Variable Forgetting Factor
V.K. Gupta, Mahesh Chandra and S.N. Sharan
Corresponding Author: V.K. Gupta
Submitted: March 10, 2012
Accepted: March 30, 2012
Published: September 01, 2012 |
Abstract:
|
In this study RLS algorithm with Double Log-Sigmoid function (DSRLS) is proposed to minimize
the effect of noise from speech signals. The performance of DSRLS is compared with the performance of RLS
and RLS algorithm with Log-Sigmoid function (SRLS). Experiments were performed on noisy data which was
prepared by adding machine gun, F16 and speech noise to clean speech samples at -5dB, 0dB, 5dB and 10dB
SNR levels. The simulation results show that both SRLS and DSRLS perform better than RLS in terms of SNR
improvement. However, DSRLS performs best in terms of SNR improvement with MSE decrement.
Key words: Adaptive filter, DSRLS, mean square error (MSE), signal to noise ratio (SNR), SRLS, VFFRLS,
|
|
Abstract
|
PDF
|
HTML |
|
Cite this Reference:
V.K. Gupta, Mahesh Chandra and S.N. Sharan, . Noise Minimization from Speech Signals using RLS Algorithm with Variable Forgetting Factor. Research Journal of Applied Sciences, Engineering and Technology, (17): 3102-3107.
|
|
|
|
 |
ISSN (Online): 2040-7467
ISSN (Print): 2040-7459 |
 |
| Information |
|
|
|
|
| Sales & Services |
|
|
|
|
