An efficient speech recognition system in adverse conditions using the nonparametric regression

General Regression Neural Networks (GRNN) have been applied to phoneme identification and isolated word recognition in clean speech. In this paper, the authors extended this approach to Arabic spoken word recognition in adverse conditions. In fact, noise robustness is one of the most challenging problems in Automatic Speech Recognition (ASR) and most of the existing recognition methods, which have shown to be highly efficient under noise-free conditions, fail drastically in noisy environments. The proposed system was tested for Arabic digit recognition at different Signal-to-Noise Ratio (SNR) levels and under four noisy conditions: multispeakers babble background, car production hall (factory), military vehicle (leopard tank) and fighter jet cockpit (buccaneer) issued from NOISEX-92 database. The proposed scheme was successfully compared to the similar recognizers based on the Multilayer Perceptrons (MLP), the Elman Recurrent Neural Network (RNN) and the discrete Hidden Markov Model (HMM). The experimental results showed that the use of nonparametric regression with an appropriate smoothing factor (spread) improved the generalization power of the neural network and the global performance of the speech recognizer in noisy environments.     

Numerical investigation of low Prandtl number effect on mixed convection in a horizontal channel by the lattice Boltzmann method

The main purpose of this study is to numerically investigate the Prandtl number effect on mixed convection in a horizontal channel heated from below using the thermal lattice Boltzmann method (TLBM). The double-population model with two different lattices is used, in particular, the D2Q9 for the velocity field and D2Q5 for the thermal field. The developed lattice Boltzmann method code to simulate the fluid flow and heat transfer in the channel was validated with available literature results based on classical numerical methods, especially the finite volume method for Pr = 6.4 and the finite difference method for Pr = 0.667. The results obtained with the TLBM have shown good agreement with the conventional methods cited. The dynamic and thermal characteristics of the fluid flow were examined in the field of low Prandtl number, such that 0.05 ≤ Pr ≤ 0.667, and also compared to Pr = 6.4; for Ra = 2420 and 7400, the Reynolds number was fixed at 1. The results showed that the influence is relatively significant for the dynamic structure of flow convection for Pr ≤ 0.3 and is little influential beyond this value.     

J integral solution for semi-elliptical surface crack in high density poly-ethylene pipe under bending

The goal of this work is to analyse the severity of semi-elliptical crack defects and to study the degree of damage in the poly-ethylene pipe in bending during the crack propagation. The semi-elliptical cracks are considered in this work located in different position in the wall of the pipe. The three finite element method based on the computation of the J integral was used to analyse the fracture behaviour of these structures. The effect of the position, shape and size of the crack on the J integral values was highlighted. The effects of strain rate and the temperature on the J integral values were also examined. The obtained results show that the strain rates have a strong influence on the J integral values especially for circumferential crack at higher bending moment. However, the energy for circumferential crack is more important compared to axial crack. The effect of the depth of the crack becomes important when the ratio (a/t) reaches a critical value of 0.6 (a/t = 0.6), especially when the ratio a/c is weak (semi-elliptical crack, a/c = 0.2) where the J integral values becomes independently of the crack depth, this conclusion is opposite to the above for the poly-ethylene pipe subjected to internal pressure. We recall finally, that the temperature effect on circumferential cracks behaviour is more important compared to the axial cracks at critical crack size (a/c = 0.2 and a/t = 0.6). It is also shown that in the wall of pipe, the internal cracks are more dangerous than the external cracks.     

Characterisation of whey protein isolate–gum tragacanth electrostatic interactions in aqueous solutions

The main objectives of this study were to measure molecular parameters of gum tragacanth by GPC‐MALLS system and investigate the complexation behaviour of whey protein isolate/gum tragacanth mixed dispersions (0.5 wt% total biopolymer concentration) as a function of pH (7.00–2.00) and the biopolymer mixing ratio (r = 0.1–10) using spectrophotometric, zeta potential and precipitate yield determination methods. GPC‐MALLS revealed that gum tragacanth contains relatively heterogeneous particles with high weight‐average and number‐average (M_w = 7.74 × 10⁵ g mol^?1 and M_n = 3.87 × 10⁵ g mol^?1) molecular mass and high dispersity index (~2.04 ± 0.3). Results of complexation displayed that as the biopolymer mixing ratio increases, the net neutrality shifts to the higher pHs. The critical values associated with the complex structure formation were found at r = 2 in which the charge density of the mixture was near zero at a wide range of pH (3.0–4.0). However, the highest precipitate yield achieved in pH 3.4.     

Automatic segmentation of newborns’ skull and fontanel from CT data using model-based variational level set

The newborn’s cranium is composed of flat cranial bone and fontanels forming together the envelope of the cerebral cavity. The fontanels are relatively flexible since they consist of fibrous membrane that ossifies during maturation becoming flat cranial bone as well. Fontanels give less contrast in computerized tomography (CT) images; they can be identified as gaps between the cranial bones. In this paper, we propose an automatic model-based method using variational level set to segment the skull and fontanels from CT images. In this approach, firstly a skull model consisting of cranial bones and fontanels is created and then used as constraint for level set evolution. Then, by removing the cranial bones from the segmented skulls, the fontanels are obtained. To verify the validity of the achieved results, automatically segmented skull and fontanels have been compared with the ones manually segmented by an expert using Dice similarity and Hausdorff dissimilarity measures, which show the good agreement between them. Furthermore, the surface areas of cranium and fontanel have been determined for these segmentations. The results for both, manual and automatic segmentation, are in good agreement.     

Properties of ordinary concretes incorporating crushed queen scallop shells

The large quantities of seashells produced in the North-West of France and the depletion of quarries motivates the research of new sources of aggregates. This article examines the influence of incorporating crushed queen scallop shells on ordinary concretes in order to assess the feasibility of reusing marine by products in the form of aggregates. For this purpose four different concrete mixes were produced, one of them used as reference with natural aggregates and the others prepared with 20, 40 and 60 % of crushed queen scallop shells replacing the natural aggregates. Concretes were characterized at fresh state by its workability, the air content and the density, and at hardened state by the mechanical and durability properties. The incorporation of crushed queen scallops causes an increase of the entrapped air of concrete, a reduction of the mechanical properties and an increase of the concrete porosity facilitating the transport of fluids and chloride ions in concrete.     

Negative Binomial Quasi‐Likelihood Inference for General Integer‐Valued Time Series Models

Two negative binomial quasi‐maximum likelihood estimates (NB‐QMLEs) for a general class of count time series models are proposed. The first one is the profile NB‐QMLE calculated while arbitrarily fixing the dispersion parameter of the negative binomial likelihood. The second one, termed two‐stage NB‐QMLE, consists of four stages estimating both conditional mean and dispersion parameters. It is shown that the two estimates are consistent and asymptotically Gaussian under mild conditions. Moreover, the two‐stage NB‐QMLE enjoys a certain asymptotic efficiency property provided that a negative binomial link function relating the conditional mean and conditional variance is specified. The proposed NB‐QMLEs are compared with the Poisson QMLE asymptotically and in finite samples for various well‐known particular classes of count time series models such as the Poisson and negative binomial integer‐valued GARCH model and the INAR(1) model. Application to a real dataset is given.     

Effect of fatigue damage on the dynamic tensile behavior of 6061-T6 aluminum alloy and AISI 4140T steel

Usually material properties are determined from damage free materials, but it is not well known how these properties vary with respect to previous fatigue damage. In the present work the dynamic response of fatigue damaged 6061-T6 aluminum alloy and AISI 4140T steel specimens subjected to impact loading was investigated. Samples subjected to previous damage under high cycle fatigue and low cycle fatigue were tested. Different fatigue damage levels were considered. In addition, the effect of previous fatigue damage on the quasi-static behavior, ductility and fracture mechanism was also evaluated for both materials. A tensile Hopkinson bar apparatus was used in this work to investigate the dynamic response of the pre-fatigued specimens. Projectile speeds ranged from 18 m/s to 30 m/s giving strains rates from 550 to 2850 s⁻¹. The quasi-static mechanical properties of aluminum are not affected by the way the fatigue damage is induced. The dynamic properties, however, are sensitive to the previous fatigue damage, but are not affected by the strain rate. In the steel case, and when damage is induced by strain control, the mechanical properties are influenced by the previous fatigue damage. The dynamic properties are sensitive to the previous fatigue damage and depend on the strain rates. The analysis results show an increase in the ductility of the aluminum alloy when increasing the fatigue damage level; the steel exhibits an opposite behavior, a decrease in the ductility when increasing the damage level. The results show how the previous fatigue damage can modify the quasi-static and dynamic mechanical properties of the tested materials.