Comparison between statistical and multiresolution wavelets analysis methods for speech classifications

Article type: Research Article

Authors: Al-Assaf, Yousef

Affiliations: American University of Sharjah, P.O. Box 26666, Sharjah, United Arab Emirates. E-mail: yassaf@aus.ac.ae

Abstract: This paper considers the use of MultiResolution Wavelets Analysis (MRWA) to classify speech for multifunctional control applications. While the classification problem is the focus of this paper, the ultimate goal is to improve speech analysis and their application performances. The proposed method performance was compared to that of temporal-time domain features with ANN classification. Results have shown that MRWA resulted in higher recognition rate and consistent output demonstrating higher reliability.

Keywords: speech analysis, classification, neural networks, wavelets analysis

Journal: Journal of Intelligent & Fuzzy Systems, vol. 14, no. 1, pp. 49-58, 2003