This paper presents an HMM-driven hybrid speech synthesis approach in which unit selection concatenative synthesis is used to improve the quality of the statistical system using a Local Minimum Generation Error (LMGE) during the synthesis stage. The idea behind this approach is to combine the robustness due to HMMs with the naturalness of concatenated units. Unlike the conventional hybrid appro...

A sufficient condition for the energy of a point such that a local minimum of the energy exists at every triangular lattice point is obtained. The condition is expressed as a certain type of strong convexity condition of the function which defines the energy. New results related to Riemann sum of a function with such the convexity and new inequalities related to sums on triangular lattice point...

It is well known that sparse matrix factorizations suffer fill. That is, some of the zero entries in a sparse matrix will become nonzero during factorization. To reduce factorization time and storage, it is important to arrange the computation so that the amount of fill is kept small. It is also well known that the amount of fill is often influenced greatly by how the rows and columns of the sp...

We discuss here the convergence of relaxation methods for MPCC with approximate sequence of stationary points by presenting a general framework to study these methods. It has been pointed out in the literature, [25], that relaxation methods with approximate stationary points fail to give guarantee of convergence. We show that by de ning a new strong approximate stationarity we can attain the de...

High-resolution LEED (low-energy electron diffraction) data of Cu(001) reveal an uniaxial in-plane lattice reconstruction by 1%. One-dimensional nanogrooves induced by ion bombardment involve the creation of steps that enable this reconstruction. This is the first verification of van der Merwe's prediction of step facilitated reconstruction. We confirm the predicted dependence on step orientati...