We propose a method to transform the on line speech signal so as to comply with the specications of an HMM-based automatic speech recognizer. The spectrum of the input signal undergoes a vocal tract length (VTL) normalization based on dierences of the average third formant F3. The high frequency gap which is generated after scaling is estimated by means of an extrapolation scheme. Mel scale c...

We study the space of linear di↵erence equations with periodic coe cients and (anti)periodic solutions. We show that this space is isomorphic to the space of tame frieze patterns and closely related to the moduli space of configurations of points in the projective space. We define the notion of combinatorial Gale transform which is a duality between periodic di↵erence equations of di↵erent orde...

image denoising by block matching and threedimensionaltransform filtering (bm3d) is a two steps state-ofthe-art algorithm that uses the redundancy of similar blocks innoisy image for removing noise. similar blocks which can havesome overlap are found by a block matching method and groupedto make 3-d blocks for 3-d transform filtering. in this paper wepropose a new block grouping algorithm in th...

It has been demonstrated that higher order recurrent neural networks exhibit an underlying fractal attractor as an artifact of their dynamics. These fractal attractors o er a very e cent mechanism to encode visual memories in a neural substrate, since even a simple twelve weight network can encode a very large set of di erent images. The main problem in this memory model, which so far has remai...

we have developed a three level implicit method for solution of the helmholtz equation. using the cubic spline in space and nite dierence in time directions. the approach has been modied to drive numerov type nite dierence method. the method yield the tri- diagonal linear system of algebraic equations which can be solved by using a tri-diagonal solver. stability and error estimation of the...

This thesis deals with fractal compression of still images based on the theory of iterated function systems IFS After an overview of the main lossy and lossless still image compression methods we introduce the IFS theory Then we detail the main fractal compression algorithms proposed in the literature A coder involves compari sons between two sets of blocks A block in the partition of the origi...

We present a novel method for the automated detection of fragments showing dissimilar expression in mRNA di erential display. The analysis is based on aligning the numerical electrophoretic lane data in respect of a given distance function de ned on a set of fragments, or signal peaks in general. We presume that signi cant dissimilarities between peaks result in extreme score values computed fo...

The aim of this study is the evaluation of the e ect of microencapsulation of nanoparticles in composite microparticles on the reduction of burst release. Microparticles (simple and composite) and nanoparticles were prepared by using water-in-oil-in-water (W/O1/W2 double-emulsion solvent di usion/evaporation method), using di erent drug/polymer ratios. For preparation of the composite micropart...

In this paper, the reduced differential transform method (RDTM) is applied to various nonlinear evolution equations, Korteweg–de Vries Burgers' (KdVB) equation, Drinefel’d–Sokolov–Wilson equations, coupled Burgers equations and modified Boussinesq equation. Approximate solutions obtained by the RDTM are compared with the exact solutions. The present results are in good agreement with the exact ...

Submitted: May 27, 2013; Accepted: Jul 5, 2013; Published: Jul 22, 2013 Abstract: In this paper, we present an algorithm called the Reduced Differential Transform Method (RDTM) to obtain approximate solutions for the Fitzhugh-Nagumo (FN) equation, Ito equation and find an exact solution to nonlinear PDE. The numerical results show that this method is a powerful tool for solving nonlinear PDEs a...