Improved Spectral Subtraction with Time Recursive Noise Estimation
P.Sunitha1, K.Satya Prasad2
1P.Sunitha, Research Scholar, Department of ECE, Jawaharlal Nehru Technological University, Kakinada (Andhra Pradesh), India.
2K.Satya Prasad, Rector, VFSTR, Vignan’s Foundation for Science, Technology & Research, Guntur (Andhra Pradesh), India.
Manuscript received on 18 June 2019 | Revised Manuscript received on 25 June 2019 | Manuscript published on 30 June 2019 | PP: 132-137 | Volume-8 Issue-5, June 2019 | Retrieval Number: D6830048419/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. 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: His Paper proposes a method for single channel speech enhancement with noise estimation algorithm in presence of additive back ground noise which overcomes the drawbacks of basic spectral subtraction. Since speech is non-Stationary signal, noise distribution is non-uniform i.e few frequency components are affected severely than others. So speech enhancementalgorithms requires an accurate noise estimation to remove the noise effectively. The proposed method uses noise estimation by First order recursive averaging algorithm based on a posteriori Signal -to-Noise-Ratio and its performance evaluated in terms of objective performance measures such as segmental SNR, Cepstrum distance, Log Likelihood Ratio and Perceptual Evaluation of Speech quality under eight different real-world noises at three ranges of input SNR. The performance of Proposed method was compared with Basic Spectral Subtraction method ,Spectral subtraction with Various noise estimation algorithms .The results shows that ,proposed method shows superior performance in all the cases considered.
Keywords: Peech Enhancement, Voice Activity Detection, Noise Power Estimation, Signal To Noise, Ratio, Objective Performance Measures.
Scope of the Article: Performance Evaluation