Exploiting multiple feature sets in data-driven impostor dataset selection for speaker verification

This study assesses the recently proposed data-driven background dataset refinement technique for speaker verification using alternate SVM feature sets to the GMM supervector features for which it was originally designed. The performance improvements brought about in each trialled SVM configuration demonstrate the versatility of background dataset refinement. This work also extends on the originally proposed technique to exploit support vector coefficients as an impostor suitability metric in the data-driven selection process. Using support vector coefficients improved the performance of the refined datasets in the evaluation of unseen data. Further, attempts are made to exploit the differences in impostor example suitability measures from varying features spaces to provide added robustness.

Darjeeling, a feature-rich VM for the resource poor

The programming and retasking of sensor nodes could benefit greatly from the use of a virtual machine (VM) since byte code is compact, can be loaded on demand, and interpreted on a heterogeneous set of devices. The challenge is to ensure good programming tools and a small footprint for the virtual machine to meet the memory constraints of typical WSN platforms. To this end we propose Darjeeling, a virtual machine modelled after the Java VM and capable of executing a substantial subset of the Java language, but designed specifically to run on 8- and 16-bit microcontrollers with 2 - 10 KB of RAM. The Darjeeling VM uses a 16- rather than a 32-bit architecture, which is more efficient on the targeted platforms. Darjeeling features a novel memory organisation with strict separation of reference from non-reference types which eliminates the need for run-time type inspection in the underlying compacting garbage collector. Darjeeling uses a linked stack model that provides light-weight threads, and supports synchronisation. The VM has been implemented on three different platforms and was evaluated with micro benchmarks and a real-world application. The latter includes a pure Java implementation of the collection tree routing protocol conveniently programmed as a set of cooperating threads, and a reimplementation of an existing environmental monitoring application. The results show that Darjeeling is a viable solution for deploying large-scale heterogeneous sensor networks. Copyright 2009 ACM.

EPR and optical absorption spectral studies on voglite mineral

A voglite mineral sample of Volrite Canyon #1 mine, Frey Point, White Canyon Mine District, San Juan County, Utah, USA is used in the present study. An EPR study on powdered sample confirms the presence of Mn(II) and Cu(II). Optical absorption spectral results are due to Cu(II) which is in distorted octahedron. NIR results are indicating the presence of water fundamentals.

Spectroscopy of selected copper group minerals : chalcophyllite and chenevixite-implications for hydrogen bonding

NIR and IR spectroscopy has been applied for detection of chemical species and the nature of hydrogen bonding in arsenate complexes. The structure and spectral properties of copper(II) arsenate minerals chalcophyllite and chenevixite are compared with copper(II) sulphate minerals devilline, chalcoalumite and caledonite. Split NIR bands in the electronic spectrum of two ranges 11700-8500 cm-1 and 8500-7200 cm-1 confirm distortion of octahedral symmetry for Cu(II) in the arsenate complexes. The observed bands with maxima at 9860 and 7750 cm-1 are assigned to Cu(II) transitions 2B1g ® 2B2g and 2B1g ® 2A1g. Overlapping bands in the NIR region 4500-4000 cm-1 is the effect of multi anions OH-, (AsO4)3- and (SO4)2-. The observation of broad and diffuse bands in the range 3700-2900 cm-1 confirms strong hydrogen bonding in chalcophyllite relative to chenevixite. The position of the water bending vibrations indicates the water is strongly hydrogen bonded in the mineral structure. The strong absorption feature centred at 1644 cm-1 in chalcophyllite indicates water is strongly hydrogen bonded in the mineral structure. The H2O-bending vibrations shift to low wavenumbers in chenevixite and an additional band observed at 1390 cm-1 is related to carbonate impurity. The characterisation of IR spectra by ν3 antisymmetric stretching vibrations of (SO4)2- and (AsO4)3 ions near 1100 and 800 cm-1 respectively is the result of isomorphic substitution for arsenate by sulphate in both the minerals of chalcophyllite and chenevixite.

Wide-angle visual feature matching for outdoor localization

Wide-angle images exhibit significant distortion for which existing scale-space detectors such as the scale-invariant feature transform (SIFT) are inappropriate. The required scale-space images for feature detection are correctly obtained through the convolution of the image, mapped to the sphere, with the spherical Gaussian. A new visual key-point detector, based on this principle, is developed and several computational approaches to the convolution are investigated in both the spatial and frequency domain. In particular, a close approximation is developed that has comparable computation time to conventional SIFT but with improved matching performance. Results are presented for monocular wide-angle outdoor image sequences obtained using fisheye and equiangular catadioptric cameras. We evaluate the overall matching performance (recall versus 1-precision) of these methods compared to conventional SIFT. We also demonstrate the use of the technique for variable frame-rate visual odometry and its application to place recognition.

Polydipsia : a feature of peritoneal dialysis

Background Some dialysis patients fail to comply with their fluid restriction causing problems due to volume overload. These patients sometimes blame excessive thirst. There has been little work in this area and no work documenting polydipsia among peritoneal dialysis (PD) patients. Methods We measured motivation to drink and fluid consumption in 46 haemodialysis patients (HD), 39 PD patients and 42 healthy controls (HC) using a modified palmtop computer to collect visual analogue scores at hourly intervals. Results Mean thirst scores were markedly depressed on the dialysis day (day 1) for HD (P<0.0001). The profile for day 2 was similar to that of HC. PD generated consistently higher scores than HD day 1 and HC (P = 0.01 vs. HC and P<0.0001 vs HD day 1). Reported mean daily water consumption was similar for HD and PD with both significantly less than HC (P<0.001 for both). However, measured fluid losses were similar for PD and HC whilst HD were lower (P<0.001 for both) suggesting that the PD group may have underestimated their fluid intake. Conclusion Our results indicate that HD causes a protracted period of reduced thirst but that the population's thirst perception is similar to HC on the interdialytic day despite a reduced fluid intake. In contrast, the PD group recorded high thirst scores throughout the day and were apparently less compliant with their fluid restriction. This is potentially important because the volume status of PD patients influences their survival.

Spectral coding methods for speech compression and speaker identification

This thesis investigates aspects of encoding the speech spectrum at low bit rates, with extensions to the effect of such coding on automatic speaker identification. Vector quantization (VQ) is a technique for jointly quantizing a block of samples at once, in order to reduce the bit rate of a coding system. The major drawback in using VQ is the complexity of the encoder. Recent research has indicated the potential applicability of the VQ method to speech when product code vector quantization (PCVQ) techniques are utilized. The focus of this research is the efficient representation, calculation and utilization of the speech model as stored in the PCVQ codebook. In this thesis, several VQ approaches are evaluated, and the efficacy of two training algorithms is compared experimentally. It is then shown that these productcode vector quantization algorithms may be augmented with lossless compression algorithms, thus yielding an improved overall compression rate. An approach using a statistical model for the vector codebook indices for subsequent lossless compression is introduced. This coupling of lossy compression and lossless compression enables further compression gain. It is demonstrated that this approach is able to reduce the bit rate requirement from the current 24 bits per 20 millisecond frame to below 20, using a standard spectral distortion metric for comparison. Several fast-search VQ methods for use in speech spectrum coding have been evaluated. The usefulness of fast-search algorithms is highly dependent upon the source characteristics and, although previous research has been undertaken for coding of images using VQ codebooks trained with the source samples directly, the product-code structured codebooks for speech spectrum quantization place new constraints on the search methodology. The second major focus of the research is an investigation of the effect of lowrate spectral compression methods on the task of automatic speaker identification. The motivation for this aspect of the research arose from a need to simultaneously preserve the speech quality and intelligibility and to provide for machine-based automatic speaker recognition using the compressed speech. This is important because there are several emerging applications of speaker identification where compressed speech is involved. Examples include mobile communications where the speech has been highly compressed, or where a database of speech material has been assembled and stored in compressed form. Although these two application areas have the same objective - that of maximizing the identification rate - the starting points are quite different. On the one hand, the speech material used for training the identification algorithm may or may not be available in compressed form. On the other hand, the new test material on which identification is to be based may only be available in compressed form. Using the spectral parameters which have been stored in compressed form, two main classes of speaker identification algorithm are examined. Some studies have been conducted in the past on bandwidth-limited speaker identification, but the use of short-term spectral compression deserves separate investigation. Combining the major aspects of the research, some important design guidelines for the construction of an identification model when based on the use of compressed speech are put forward.

Coding speech signals at very low rates (below 1 kb/s) with high intelligibility

This thesis presents an original approach to parametric speech coding at rates below 1 kbitsjsec, primarily for speech storage applications. Essential processes considered in this research encompass efficient characterization of evolutionary configuration of vocal tract to follow phonemic features with high fidelity, representation of speech excitation using minimal parameters with minor degradation in naturalness of synthesized speech, and finally, quantization of resulting parameters at the nominated rates. For encoding speech spectral features, a new method relying on Temporal Decomposition (TD) is developed which efficiently compresses spectral information through interpolation between most steady points over time trajectories of spectral parameters using a new basis function. The compression ratio provided by the method is independent of the updating rate of the feature vectors, hence allows high resolution in tracking significant temporal variations of speech formants with no effect on the spectral data rate. Accordingly, regardless of the quantization technique employed, the method yields a high compression ratio without sacrificing speech intelligibility. Several new techniques for improving performance of the interpolation of spectral parameters through phonetically-based analysis are proposed and implemented in this research, comprising event approximated TD, near-optimal shaping event approximating functions, efficient speech parametrization for TD on the basis of an extensive investigation originally reported in this thesis, and a hierarchical error minimization algorithm for decomposition of feature parameters which significantly reduces the complexity of the interpolation process. Speech excitation in this work is characterized based on a novel Multi-Band Excitation paradigm which accurately determines the harmonic structure in the LPC (linear predictive coding) residual spectra, within individual bands, using the concept 11 of Instantaneous Frequency (IF) estimation in frequency domain. The model yields aneffective two-band approximation to excitation and computes pitch and voicing with high accuracy as well. New methods for interpolative coding of pitch and gain contours are also developed in this thesis. For pitch, relying on the correlation between phonetic evolution and pitch variations during voiced speech segments, TD is employed to interpolate the pitch contour between critical points introduced by event centroids. This compresses pitch contour in the ratio of about 1/10 with negligible error. To approximate gain contour, a set of uniformly-distributed Gaussian event-like functions is used which reduces the amount of gain information to about 1/6 with acceptable accuracy. The thesis also addresses a new quantization method applied to spectral features on the basis of statistical properties and spectral sensitivity of spectral parameters extracted from TD-based analysis. The experimental results show that good quality speech, comparable to that of conventional coders at rates over 2 kbits/sec, can be achieved at rates 650-990 bits/sec.

A new approach for finding an appropriate combination of texture parameters for classification

This paper suggests an approach for finding an appropriate combination of various parameters for extracting texture features (e.g. choice of spectral band for extracting texture feature, size of the moving window, quantization level of the image, and choice of texture feature etc.) to be used in the classification process. Gray level co-occurrence matrix (GLCM) method has been used for extracting texture from remotely sensed satellite image. Results of the classification of an Indian urban environment using spatial property (texture), derived from spectral and multi-resolution wavelet decomposed images have also been reported. A multivariate data analysis technique called ‘conjoint analysis’ has been used in the study to analyze the relative importance of these parameters. Results indicate that the choice of texture feature and window size have higher relative importance in the classification process than quantization level or the choice of image band for extracting texture feature. In case of texture features derived using wavelet decomposed image, the parameter ‘decomposition level’ has almost equal relative importance as the size of moving window and the decomposition of images up to level one is sufficient and there is no need to go for further decomposition. It was also observed that the classification incorporating texture features improves the overall classification accuracy in a statistically significant manner in comparison to pure spectral classification.