Thermoelectric Properties of Co4Sb12 Skutterudite Materials with Partial In Filling and Excess In Additions

The properties of Co4Sb12 with various In additions were studied. X-ray diffraction revealed the presence of the pure δ-phase of In0.16Co4Sb12, whereas impurity phases (γ-CoSb2 and InSb) appeared for x = 0.25, 0.40, 0.80, and 1.20. The homogeneity and morphology of the samples were observed by Seebeck microprobe and scanning electron microscopy, respectively. All the quenched ingots from which the studied samples were cut were inhomogeneous in the axial direction. The temperature dependence of the Seebeck coefficient (S), electrical conductivity (σ), and thermal conductivity (κ) was measured from room temperature up to 673 K. The Seebeck coefficient of all In-added Co4Sb12 materials was negative. When the filler concentration increases, the Seebeck coefficient decreases. The samples with In additions above the filling limit (x = 0.22) show an even lower Seebeck coefficient due to the formation of secondary phases: InSb and CoSb2. The temperature variation of the electrical conductivity is semiconductor-like. The thermal conductivity of all the samples decreases with temperature. The central region of the In0.4Co4Sb12 ingot shows the lowest thermal conductivity, probably due to the combined effect of (a) rattling due to maximum filling and (b) the presence of a small amount of fine-dispersed secondary phases at the grain boundaries. Thus, regardless of the non-single-phase morphology, a promising ZT (S 2 σT/κ) value of 0.96 at 673 K has been obtained with an In addition above the filling limit.

Influence of Spatial Scales on the Performance of Saturated Subsurface Flow and Transport Models

Numerical modeling of saturated subsurface flow and transport has been widely used in the past using different numerical schemes such as finite difference and finite element methods. Such modeling often involves discretization of the problem in spatial and temporal scales. The choice of the spatial and temporal scales for a modeling scenario is often not straightforward. For example, a basin-scale saturated flow and transport analysis demands larger spatial and temporal scales than a meso-scale study, which in turn has larger scales compared to a pore-scale study. The choice of spatial-scale is often dictated by the computational capabilities of the modeler as well as the availability of fine-scale data. In this study, we analyze the impact of different spatial scales and scaling procedures on saturated subsurface flow and transport simulations.

Joint pitch-analysis formant-synthesis framework for CS recovery of speech

A joint analysis-synthesis framework is developed for the compressive sensing (CS) recovery of speech signals. The signal is assumed to be sparse in the residual domain with the linear prediction filter used as the sparse transformation. Importantly this transform is not known apriori, since estimating the predictor filter requires the knowledge of the signal. Two prediction filters, one comb filter for pitch and another all pole formant filter are needed to induce maximum sparsity. An iterative method is proposed for the estimation of both the prediction filters and the signal itself. Formant prediction filter is used as the synthesis transform, while the pitch filter is used to model the periodicity in the residual excitation signal, in the analysis mode. Significant improvement in the LLR measure is seen over the previously reported formant filter estimation.

Automatic speech segmentation using probabilistic latent component modeling

Latent variable methods, such as PLCA (Probabilistic Latent Component Analysis) have been successfully used for analysis of non-negative signal representations. In this paper, we formulate PLCS (Probabilistic Latent Component Segmentation), which models each time frame of a spectrogram as a spectral distribution. Given the signal spectrogram, the segmentation boundaries are estimated using a maximum-likelihood approach. For an efficient solution, the algorithm imposes a hard constraint that each segment is modelled by a single latent component. The hard constraint facilitates the solution of ML boundary estimation using dynamic programming. The PLCS framework does not impose a parametric assumption unlike earlier ML segmentation techniques. PLCS can be naturally extended to model coarticulation between successive phones. Experiments on the TIMIT corpus show that the proposed technique is promising compared to most state of the art speech segmentation algorithms.