Sharif Digital Repository

An extension to conventional Hidden Markov Model (HMM)-based speech enhancement method is developed. An algebraic method is proposed to estimate gain of speech and noise in order to improve the quality of the estimated speech. Different pronunciations and intonations may affect speech gain. Besides, gain of noise may vary remarkably from one environment to the other one. This may lead in a mismatch between energy contour of trained models and energy contour of noisy speech signal. In this work, speech gain and noise gain are estimated based on an algebraic method simultaneously in order to match gain of noisy speech and noisy model. To carry out this procedure an extension of least square... 

To estimate the parameters of fuzzy linear regression models with fuzzy output and crisp inputs, we develop a mathematical programming model in this paper. The method is constructed on the basis of minimizing the square of the total difference between observed and estimated spread values or in other words minimizing the least square errors. The advantage of the proposed approach is its simplicity in programming and computation as well as its performance. To compare the performance of the proposed approach with the other methods, two examples are presented. © 2004 Elsevier Inc. All rights reserved  

In the welding process, droplet detachment regularity is important for a good welding quality. The frequency of detachment is one of influential indices for assessing the quality of GMAW in the globular transfer mode. In this paper, a relationship is estimated for measuring droplet detachment frequency in the globular GMAW. The droplet detachment frequency is estimated with a curve fitting by the least squares error (LSE) method. By using short time frequency transform (STFT) we can extract the frequency contents from the mass of drop that is one of the states in the welding model  

Weighting elements to achieve radiation patterns with desired characteristics is a classical work in phased array antennas. These characteristics can be low sidelobe level, narrow beamwidth, high directivity, pattern nulling in special angle and etc. For each of these characteristics, different methods have been introduced. Most of methods have been presented for uniform arrays, however there are lots of methods to obtain a desired pattern for antennas with non-uniform element distances. The problem with these methods is complexity or not very good results. In this paper, fast and easy methods based on Least Square Error that leads to good results are presented. In addition, weighting of...