Performance on Speech Enhancement Objective Quality Measures Using Hybrid Wavelet Thresholding
S.China Venkateswarlu1, A.Karthik2, K. Naveen Kumar3
1Dr.S.China Venkateswarlu, Professor in Dept.of ECE, Institute of Aeronautical Engineering (IARE), Dundigal, Hyderabad, T.G, India.
2A.Karthik, Assistant Professor in Dept.of ECE, Institute of Aeronautical Engineering (IARE), Dundigal, Hyderabad, T.G, India.
3Dr. K. Naveen Kumar, Professor of ECE Department, Marri Laxman Reddy Institute of Technology and Management, Dundigal, Hyderabad, India.
Manuscript received on January 20, 2019. | Revised Manuscript received on February 14 2019. | Manuscript published on August 30, 2019. | PP: 3523-3533 | Volume-8 Issue-6, August 2019. | Retrieval Number: F9343088619/19©BEIESP | DOI: 10.35940/ijeat.F9343.088619
Open Access | Ethics and Policies | Cite | Mendeley
© 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: The quality of being easily understandable of the speech signals are very importantin communication and other speech related systems. In order to improve these two in thespeech signal, Speech improvement sets of computer instructions and devices are used so that itmay be better fully used by other speech processing setsof computer instructions. Most of the speech communication that requires atleast one microphone and the desired speech signal is usually contaminated by backgroundnoise and echo. As a result, the speech sign must be “cleaned” with advanced sign preparing devices before it is played out, transmitted, or put away. In this venture it has been investigated the required things and degree of upgrades in the field of discourse improvement utilizing discourse de-noising sets of PC directions announced in books with the fundamental intend to concentrate on the utilization of the window shape limits/rules in STSA based Speech Improvement process in which the sign destroyed by commotion is into edges and each part/segment is Windowed and the Windowed Speech pieces/parts zone connected to the Speech Improvement set of PC guidelines and the Improved Speech sign is modified in its time area. In general, the Speech Improvement methods make use theHamming Window for this purpose. In this work an attempt has been made to study the effect of Window shape on the Speech. The Modified Improved thresholding is proposed by Asser Ghanbari and Mohammad Reza Karami and can be used like a hard thresholding limit with respect to the wavelet coefficients through and through worth progressively conspicuous than limit esteem and resembles an exponential capacity for the wavelet coefficients supreme worth not as much as edge esteem and is characterized.
Keywords: Speech, Windows, Wavelet, Quality, Thresholding, Speech Enhancement.