Subspace Based Speech Enhancement Using Gaussian Mixture Model

Traditional subspace based speech enhancement (SSE)methods use linear minimum mean square error (LMMSE) estimation that is optimal if the Karhunen Loeve transform (KLT) coefficients of speech and noise are Gaussian distributed. In this paper, we investigate the use of Gaussian mixture (GM) density for modeling the non-Gaussian statistics of the clean speech KLT coefficients. Using Gaussian mixture model (GMM), the optimum minimum mean square error (MMSE) estimator is found to be nonlinear and the traditional LMMSE estimator is shown to be a special case. Experimental results show that the proposed method provides better enhancement performance than the traditional subspace based methods.Index Terms: Subspace based speech enhancement, Gaussian mixture density, MMSE estimation.

A Multi-layer Perceptron based Non-linear Mixture Model to estimate class abundance from mixed pixels

Sub-pixel classification is essential for the successful description of many land cover (LC) features with spatial resolution less than the size of the image pixels. A commonly used approach for sub-pixel classification is linear mixture models (LMM). Even though, LMM have shown acceptable results, pragmatically, linear mixtures do not exist. A non-linear mixture model, therefore, may better describe the resultant mixture spectra for endmember (pure pixel) distribution. In this paper, we propose a new methodology for inferring LC fractions by a process called automatic linear-nonlinear mixture model (AL-NLMM). AL-NLMM is a three step process where the endmembers are first derived from an automated algorithm. These endmembers are used by the LMM in the second step that provides abundance estimation in a linear fashion. Finally, the abundance values along with the training samples representing the actual proportions are fed to multi-layer perceptron (MLP) architecture as input to train the neurons which further refines the abundance estimates to account for the non-linear nature of the mixing classes of interest. AL-NLMM is validated on computer simulated hyperspectral data of 200 bands. Validation of the output showed overall RMSE of 0.0089±0.0022 with LMM and 0.0030±0.0001 with the MLP based AL-NLMM, when compared to actual class proportions indicating that individual class abundances obtained from AL-NLMM are very close to the real observations.

Executable Analysis using Abstract Interpretation with Circular Linear Progressions

We propose a new abstract domain for static analysis of executable code. Concrete states are abstracted using circular linear progressions (CLPs). CLPs model computations using a finite word length as is seen in any real life processor. The finite abstraction allows handling overflow scenarios in a natural and straight-forward manner. Abstract transfer functions have been defined for a wide range of operations which makes this domain easily applicable for analyzing code for a wide range of ISAs. CLPs combine the scalability of interval domains with the discreteness of linear congruence domains. We also present a novel, lightweight method to track linear equality relations between static objects that is used by the analysis to improve precision. The analysis is efficient, the total space and time overhead being quadratic in the number of static objects being tracked.

Fluctuation quench and hydrophobic collapse of polymers and biopolymers in water-DMSO binary mixture at low DMSO concentration

Binary mixtures have strong influence on activities of polymers and biopolymers even at low cosolvent concentration. Among the several aqueous binary mixtures studied, water-DMSO especially stands out for its unusual behavior at certain specific concentrations of DMSO. In the present work, we study the effect of water-DMSO binary mixture on polymers and biopolymers by taking a simple linear hydrocarbon chain of intermediate length (n = 30) and the protein lysozyme, respectively. We find that at a mole fraction of 0.05 of DMSO (x(DMSO) = 0.05) in aqueous solution, the hydrocarbon chain adopts the collapsed conformation as the most stable and rigid state. In this case of 0.05 mole fraction of DMSO in bulk, the DMSO concentration in the first hydration layer around the polymer is found to be as large as 17%. Formation of such hydrophobic environment around the polymer is the reason for the collapsed state gaining so much stability. Interestingly, similar quench of conformational fluctuation is also observed for the protein investigated. It is observed that in the case of alkane polymer chains, long wavelength fluctuation gets easily quenched, the polymer being purely hydrophobic. However, in case of the protein, quench of fluctuation is prominent only at the hydrophobic surface, and quench of long wavelength fluctuation becomes insignificant for the full protein. As protein contains both hydrophobic and hydrophilic moieties, the extent of quench of conformational fluctuation with respect to that in pure water is almost half for the biopolymer complex (16.83%) than the same for pure hydrophobic polymer chain (32.43%).

A thermodynamic interpretation of the zonal arrangement of minerals in the bedded manganese deposits of the Noda tamagawa mine, Japan

Calculated phase relations in the system MnOSi02-C02-02 were used to propose a thermodynamic explanation for the thermal metamorphism of rhodochrosite beds lying between chert strata. The metamorphic MnOS i 0 2 minerals are arranged in order quartz(chert), rhodonite. tephroite and manganosite-hausmannite-pyrochroite rhodonite across the ore bed. The calculation covered temperatures up to 1000 K and pressures up to 5 kb. The zoning was interpreted as the result of a continuous rise in metamorphic temperature. The equilibrium partner of rhodochrosite changed from rhodonite through manganosite. Across the ore bed there are gradients in the chemical potential of MnO and SiO2 but fugacities of volatlle components such as C02. 02 and H20 were probably uniform at any given time and location during formation of the zones. Assuming that the total pressure and the fugacity of C02 were at 1.4 kb and 1.0 1 b. respectively. rhodonite. tephroite and manganosite would have formed at 472. 478 and 629 K.

Influence of ethanol/water mixture on the undrained shear strength of pure clays

The mechanical properties of clays are highly dependent not only on the stress/strain ratio to which the material is subjected but also on the chemistry of the pore fluids which in turn affects the intergranular or the effective stresses. Atterberg limits and vane shear tests were performed with different pore fluids in order to observe how the fine-grained material mechanically responded. The diffuse double layer theory has been used to interpret the data of vane shear tests in order to explain the variation of geotechnical responses with the different clays. Van der Waals forces and double layer forces were obtained and capillary forces calculated. The results show that while for kaolinite and illite the chemistry of the pore fluids has no influence on the water content and hence on the mechanical behaviour of the material, Na-smectite shows a strong correlation between the dielectric constant of the pore fluids and an increase in undrained shear strength. The data obtained extends an understanding of the influence of the dielectric constant (epsilon) of the pore fluids on the geotechnical properties of fine-grained materials.

The irradiation of 1:1 mixture of ammonia:carbon dioxide ice at 30 K using 1 kev electrons

In this Letter the results of an experimental investigation of 1 keV electron irradiation of a 1:1 ice mixture of NH3:CO2 at 30 K was made under ultrahigh vacuum (10(-9) mbar) conditions. Molecular products formed within the ice were detected and monitored using FTIR spectroscopy. The formation of ammonium ions (NH4+), cyanate ions (OCN-), CO was observed leading to the synthesis of ammonium carbamate (NH4NH2CO2). The consequences of these results for prebiotic chemistry in the interstellar medium and star forming regions are discussed. Crown Copyright (C) 2012 Published by Elsevier B.V. All rights reserved.

Mixture solubilities of nitrobenzoic acid isomers in supercritical carbon dioxide

The ternary solubilities of solid isomers of nitrobenzoic acid (NBA) were experimentally determined at 308, 318 and 328K over a pressure range of 12-18 MPa in supercritical carbon dioxide (SCCO2). Compared to its binary solubility, the ternary solubilities of m-NBA increased at 308 K while it decreased at 328 K. However, the ternary solubilities of p-NBA increased at all temperatures and pressures except at 13 MPa and 328K. A new model was developed by applying solution model and activity coefficient model for the ternary solubilities of pharmaceutical and non-pharmaceutical solid mixtures in terms of temperature, density and cosolute composition. The model equation involves four temperature independent constraint-free parameters. The model equation correlates the ternary solubilities of seven pharmaceutical solid mixtures along with current data with an average AARD around 9.5% and sixteen non-pharmaceutical solid mixtures with 9% AARD. (C) 2012 Elsevier B.V. All rights reserved.

A mixture model approach for formant tracking and the robustness of student's-t distribution

We address the problem of robust formant tracking in continuous speech in the presence of additive noise. We propose a new approach based on mixture modeling of the formant contours. Our approach consists of two main steps: (i) Computation of a pyknogram based on multiband amplitude-modulation/frequency-modulation (AM/FM) decomposition of the input speech; and (ii) Statistical modeling of the pyknogram using mixture models. We experiment with both Gaussian mixture model (GMM) and Student's-t mixture model (tMM) and show that the latter is robust with respect to handling outliers in the pyknogram data, parameter selection, accuracy, and smoothness of the estimated formant contours. Experimental results on simulated data as well as noisy speech data show that the proposed tMM-based approach is also robust to additive noise. We present performance comparisons with a recently developed adaptive filterbank technique proposed in the literature and the classical Burg's spectral estimator technique, which show that the proposed technique is more robust to noise.

The irradiation of pure CH3OH and 1:1 mixture of NH3:CH3OH ices at 30 K using low energy electrons

The results of an experimental investigation of 1 keV electron irradiation of ices (deposited at 30 K) of (i) pure methanol and (ii) of a 1:1 mixture of NH3:CH3OH are reported. Molecular products formed within the ice were detected and monitored using FTIR spectroscopy. The products observed were methyl formate (H3COHCO), methane (CH4), hydroxymethyl (CH2OH), formamide (HCONH2), formic acid (HCOOH), formaldehyde (H2CO), formyl radical (HCO), cyanate ion (OCN-), isocyanic acid (HNCO), carbon monoxide (CO) and carbon dioxide (CO2). The consequences of these results for prebiotic chemistry in the interstellar medium and star forming regions are discussed. Crown Copyright (C) 2012 Published by Elsevier B. V. All rights reserved.

Speeded up Gaussian mixture model algorithm for background subtraction

Adaptive Gaussian Mixture Models (GMM) have been one of the most popular and successful approaches to perform foreground segmentation on multimodal background scenes. However, the good accuracy of the GMM algorithm comes at a high computational cost. An improved GMM technique was proposed by Zivkovic to reduce computational cost by minimizing the number of modes adaptively. In this paper, we propose a modification to his adaptive GMM algorithm that further reduces execution time by replacing expensive floating point computations with low cost integer operations. To maintain accuracy, we derive a heuristic that computes periodic floating point updates for the GMM weight parameter using the value of an integer counter. Experiments show speedups in the range of 1.33 - 1.44 on standard video datasets where a large fraction of pixels are multimodal.

Study of the effect of alkali mixture on V-O bond length in Oxyfluoro Vanadate glasses using Raman spectroscopy

Raman spectroscopic study on Oxyfluoro Vanadate glasses containing various proportions of lithium fluoride and rubidium fluoride was carried out to see an effect of mixture of alkali on vanadium-oxygen (V-O) bond length. Glasses with a general formula 40V(2)O(5) - 30BaF(2) - (30 - x) LiF - xRbF (x = 0-30) were prepared. Room temperature Raman spectra of these glass samples were recorded in back scattering geometry. The data presented is in ``reduced Raman intensity'' form with maximum peak scaled to 100. We have used v = Aexp(BR), where A and B are fitting parameters, to correlate the bond length R with Raman scattering frequency v. We observed that variation in bond length and its distribution about a most probable value can be correlated to the alkali environment present in these glasses. We also observed that all rubidium environment around the network forming unit is more homogenous than all lithium environment.