A Family of Spectral Subtraction Algorithms for Tamil Speech Enhancement
Vimala.C1, V.Radha2
1Vimala.C, Department of Computer Science, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore, India.
1Dr.V.Radha, Department of Computer Science, Avinashilingam Institute for Home Science and Higher Education for Women, Coimbatore, India.
Manuscript received on February 15, 2012. | Revised Manuscript received on February 20, 2012. | Manuscript published on March 05, 2012. | PP: 397-401 | Volume-2 Issue-1, March 2012. | Retrieval Number: A0461022112/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Speech enhancement aims to improve the speech quality by using various techniques and algorithms. Over the past several years there has been considerable attention focused on the enhancement of speech degraded by several types of noise. The degradation of speech due to the presence of noise causes severe difficulties in various communication environments. Noise suppression has numerous applications like Human Computer Interaction, hands-free communications, Voice over IP (VoIP), hearing aids, teleconferencing system etc. For this issue there is always a unique need for the technique which offers expected outcome with limited complexity in implementation. Hence, in this paper a family of spectral subtraction techniques is employed for Tamil speech noise cancellation due to its simplicity. The algorithms adopted for this research work are namely basic spectral subtraction, Non linear Spectral Subtraction, MultiBand Spectral Subtraction, Minimum Mean Square Error (MMSE), and Log Spectral MMSE. All these algorithms are analyzed and implemented for two types of noises namely white and babble noise. The performances of these algorithms are estimated based on SNR and MSE measures. Based on the experimental results, the Non linear spectral subtraction algorithm provides better results than any other adopted algorithms.
Keywords: Speech enhancement, Tamil Speech, Spectral Subtraction, Non linear Spectral Subtraction, MMSE, Log Spectral MMSE, SNR and MSE.