A confidence-based approach for improving keyword hypothesis scores

The task in keyword spotting (KWS) is to hypothesise times at which any of a set of key terms occurs in audio. An important aspect of such systems are the scores assigned to these hypotheses, the accuracy of which have a significant impact on performance. Estimating these scores may be formulated as a confidence estimation problem, where a measure of confidence is assigned to each key term hypothesis. In this work, a set of discriminative features is defined, and combined using a conditional random field (CRF) model for improved confidence estimation. An extension to this model to directly address the problem of score normalisation across key terms is also introduced. The implicit score normalisation which results from applying this approach to separate systems in a hybrid configuration yields further benefits. Results are presented which show notable improvements in KWS performance using the techniques presented in this work. © 2013 IEEE.

Adapting data processing to compare model and experiment accurately: A discrete element model and magnetic resonance measurements of a 3d cylindrical fluidized bed

Discrete element modeling is being used increasingly to simulate flow in fluidized beds. These models require complex measurement techniques to provide validation for the approximations inherent in the model. This paper introduces the idea of modeling the experiment to ensure that the validation is accurate. Specifically, a 3D, cylindrical gas-fluidized bed was simulated using a discrete element model (DEM) for particle motion coupled with computational fluid dynamics (CFD) to describe the flow of gas. The results for time-averaged, axial velocity during bubbling fluidization were compared with those from magnetic resonance (MR) experiments made on the bed. The DEM-CFD data were postprocessed with various methods to produce time-averaged velocity maps for comparison with the MR results, including a method which closely matched the pulse sequence and data processing procedure used in the MR experiments. The DEM-CFD results processed with the MR-type time-averaging closely matched experimental MR results, validating the DEM-CFD model. Analysis of different averaging procedures confirmed that MR time-averages of dynamic systems correspond to particle-weighted averaging, rather than frame-weighted averaging, and also demonstrated that the use of Gaussian slices in MR imaging of dynamic systems is valid. © 2013 American Chemical Society.

100 Gb/s uncooled WDM system using conventional WDM components and advanced receiver signal processing

We demonstrate an uncooled WDM system using standard WDM components and receiver signal processing, with a different number of receivers to transmitters, to allow wide temperature drift of the transmitter lasers. A 100 Gb/s 8-wavelength demonstrator has been developed, which proves the feasibility of the approach over 25 km of SMF. © 2012 Optical Society of America.

Anisotropy preserving DTI processing

Statistical analysis of diffusion tensor imaging (DTI) data requires a computational framework that is both numerically tractable (to account for the high dimensional nature of the data) and geometric (to account for the nonlinear nature of diffusion tensors). Building upon earlier studies exploiting a Riemannian framework to address these challenges, the present paper proposes a novel metric and an accompanying computational framework for DTI data processing. The proposed approach grounds the signal processing operations in interpolating curves. Well-chosen interpolating curves are shown to provide a computational framework that is at the same time tractable and information relevant for DTI processing. In addition, and in contrast to earlier methods, it provides an interpolation method which preserves anisotropy, a central information carried by diffusion tensor data. © 2013 Springer Science+Business Media New York.

Gaussian processes for POMDP-based dialogue manager optimization

A partially observable Markov decision process (POMDP) has been proposed as a dialog model that enables automatic optimization of the dialog policy and provides robustness to speech understanding errors. Various approximations allow such a model to be used for building real-world dialog systems. However, they require a large number of dialogs to train the dialog policy and hence they typically rely on the availability of a user simulator. They also require significant designer effort to hand-craft the policy representation. We investigate the use of Gaussian processes (GPs) in policy modeling to overcome these problems. We show that GP policy optimization can be implemented for a real world POMDP dialog manager, and in particular: 1) we examine different formulations of a GP policy to minimize variability in the learning process; 2) we find that the use of GP increases the learning rate by an order of magnitude thereby allowing learning by direct interaction with human users; and 3) we demonstrate that designer effort can be substantially reduced by basing the policy directly on the full belief space thereby avoiding ad hoc feature space modeling. Overall, the GP approach represents an important step forward towards fully automatic dialog policy optimization in real world systems. © 2013 IEEE.

Gut contents of bighead carp (Aristichthys nobilis) and the processing and digestion of algal cells in the alimentary canal

Bighead carp is one of the most important freshwater filter-feeding fish of Chinese aquaculture. In recent decades, there have been a number of contradictory conclusions on the digestibility of algae by bighead carp based on the results from gut contents and digestive enzyme analysis or radiolabelled isotope techniques. Phytoplankton in the gut contents of bighead carp (cultured in a large net cage in Lake Donghu) were studied during March-May. In biomass, the dominant phytoplankters in the fore-gut contents were the centric diatom Cyclotella (average 54.5%, range 33.8-74.3%) and the dinoflagellate Cryptomonas (average 22.8%, range 6.8-55.8%). Phytoplankton in water samples were generally present in proportionate amounts in samples from the fore-guts of bighead carp. The size of most phytoplankton present in the intestine of bighead carp was between 8 and 20 mum in length. Bighead carp was also able to collect particles (as small as 5-6 mum) much smaller than their filtering net meshes, suggesting the importance of mucus in collecting small particles, Examination of the change in the integrity of Cyclotella on passage through the esophagus of bighead carp indicated that disruption of the algal cell walls is principally by the pharyngeal teeth, explaining the previous contradictory conclusions. (C) 2001 Elsevier Science B.V. All rights reserved.

Quantum information processing with superconducting qubits in a microwave field

We investigate the quantum dynamics of a Cooper-pair box with a superconducting loop in the presence of a nonclassical microwave field. We demonstrate the existence of Rabi oscillations for both single- and multiphoton processes and, moreover, we propose a new quantum computing scheme (including one-bit and conditional two-bit gates) based on Josephson qubits coupled through microwaves.

Structural properties of polycrystalline silicon films formed by pulsed rapid thermal processing

A novel pulsed rapid thermal processing (PRTP) method has been used for realizing solid-phese crystallization of amorphous silicon films prepared by plasma-enhanced chemical vapour deposit ion. The microstructure and surface morphology of the crystallized films were investigated using x-ray diffraction and atomic Force microscopy. The results indicate that PRTP is a suitable post-crystallization technique for fabricating large-area polycrystalline silicon films with good structural quality, such as large grain size, small lattice microstrain and smooth surface morphology on low-cost glass substrates.

Rapid thermal annealing processing of GaN epilayer on sapphire(0 0 0 1)

The effect of rapid thermal annealing (RTA) in a Nz ambient up to 900 degrees C has been investigated for GaN films grown on sapphire(0 0 0 1) substrates. Raman spectra, X-ray diffractometry and Hall-effect studies were performed for this purpose. The Raman spectra show the presence of the E-2 (high) mode and a shift in the wave number of this mode with respect to the annealing processing. This result suggests the presence and relaxation of residual stress due to thermal expansion misfit in the films which are confirmed by X-ray measurements and the structure quality of GaN epilayer was improved. Furthermore, the electron mobility increased at room temperature with respect to decrease of background electron concentration after RTA. (C) 1998 Elsevier Science B.V. All rights reserved.

Processing network models of energy/environment systems

Processing networks are a variant of the standard linear programming network model which are especially useful for optimizing industrial energy/environment systems. Modelling advantages include an intuitive diagrammatic representation and the ability to incorporate all forms of energy and pollutants in a single integrated linear network model. Added advantages include increased speed of solution and algorithms supporting formulation. The paper explores their use in modelling the energy and pollution control systems in large industrial plants. The pollution control options in an ethylene production plant are analyzed as an example. PROFLOW, a computer tool for the formulation, analysis, and solution of processing network models, is introduced.

yet another meta-language for programming language processing

The formal specification language LFC was designed to support formal specification acquisition. However, it is yet suited to be used as a meta-language for specifying programming language processing. This paper introduces LFC as a meta-language, and compares it with ASF+SDF, an algebraic specification formalism that can also be used to programming languages.

Optical image processing using the photoinduced anisotropy of pyrrylfulgide

A synthesized photochromic compound-pyrrylfulgide-is prepared as a thin film doped in a polymethylmethacrylate (PMMA) matrix. Under irradiation by UV light, the film converts from the bleached state into a colored state that has a maximum absorption at 635 nm and is thermally stable at room temperature. When the colored state is irradiated by a linearly polarized 650 nm laser, the film returns to the bleached state; photoinduced anisotropy is produced during this process. Application of optical image processing methods using the photoinduced anisotropy of the pyrrylfulgide/PMMA film is described. Examples in non-Fourier optical image processing, such as contrast reversal and image subtraction and summation, as well as in Fourier optical image processing, such as low-pass filtering and edge enhancement, are presented. (c) 2006 Optical Society of America.