Structured discriminative models for noise robust continuous speech recognition

Recently there has been interest in structured discriminative models for speech recognition. In these models sentence posteriors are directly modelled, given a set of features extracted from the observation sequence, and hypothesised word sequence. In previous work these discriminative models have been combined with features derived from generative models for noise-robust speech recognition for continuous digits. This paper extends this work to medium to large vocabulary tasks. The form of the score-space extracted using the generative models, and parameter tying of the discriminative model, are both discussed. Update formulae for both conditional maximum likelihood and minimum Bayes' risk training are described. Experimental results are presented on small and medium to large vocabulary noise-corrupted speech recognition tasks: AURORA 2 and 4. © 2011 IEEE.

Model-based approaches to handling additive noise in reverberant environments

Model-based approaches to handle additive and convolutional noise have been extensively investigated and used. However, the application of these schemes to handling reverberant noise has received less attention. This paper examines the extension of two standard additive/convolutional noise approaches to handling reverberant noise. The first is an extension of vector Taylor series (VTS) compensation, reverberant VTS, where a mismatch function including reverberant noise is used. The second scheme modifies constrained MLLR to allow a wide-span of frames to be taken into account and projected into the required dimensionality. To allow additive noise to be handled, both these schemes are combined with standard VTS. The approaches are evaluated and compared on two tasks, MC-WSJ-AV, and a reverberant simulated version of AURORA-4. © 2011 IEEE.

The effect of a twin tunnel on the propagation of ground-borne vibration from an underground railway

Accurate predictions of ground-borne vibration levels in the vicinity of an underground railway are greatly sought after in modern urban centres. Yet the complexity involved in simulating the underground environment means that it is necessary to make simplifying assumptions about this system. One such commonly made assumption is to ignore the effects of neighbouring tunnels, despite the fact that many underground railway lines consist of twin-bored tunnels, one for the outbound direction and one for the inbound direction. This paper presents a unique model for two tunnels embedded in a homogeneous, elastic fullspace. Each of these tunnels is subject to both known, dynamic train forces and dynamic cavity forces. The net forces acting on the tunnels are written as the sum of those tractions acting on the invert of a single tunnel, and those tractions that represent the motion induced by the neighbouring tunnel. By apportioning the tractions in this way, the vibration response of a two-tunnel system is written as a linear combination of displacement fields produced by a single-tunnel system. Using Fourier decomposition, forces are partitioned into symmetric and antisymmetric modenumber components to minimise computation times. The significance of the interactions between two tunnels is quantified by calculating the insertion gains, in both the vertical and horizontal directions, that result from the existence of a second tunnel. The insertion-gain results are shown to be localised and highly dependent on frequency, tunnel orientation and tunnel thickness. At some locations, the magnitude of these insertion gains is greater than 20 dB. This demonstrates that a high degree of inaccuracy exists in any surface vibration prediction model that includes only one of the two tunnels. This novel two-tunnel solution represents a significant contribution to the existing body of research into vibration from underground railways, as it shows that the second tunnel has a significant influence on the accuracy of vibration predictions for underground railways. © 2011 Elsevier Ltd. All rights reserved.

Triggering in a thermoacoustic system with stochastic noise

This paper explores the mechanism of triggering in a simple thermoacoustic system, the Rijke tube. It is demonstrated that additive stochastic perturbations can cause triggering before the linear stability limit of a thermoacoustic system. When triggering from low noise amplitudes, the system is seen to evolve to self-sustained oscillations via an unstable periodic solution of the governing equations. Practical stability is introduced as a measure of the stability of a linearly stable state when finite perturbations are present. The concept of a stochastic stability map is used to demonstrate the change in practical stability limits for a system with a subcritical bifurcation, once stochastic terms are included. The practical stability limits are found to be strongly dependent on the strength of noise.

The propagation of the mud crab Scylla serrata (F.) de Haan

The mud crab Scylla serrata is an important commercial species found in many brackish areas in the Philippines. During spawning and hatching, the berried females migrate to the sea. Seeds for pond stocking are obtained from the wild. Because of the unpredictability of seed supply, there is a need to propagate the species artificially. Thus, spawning, larval rearing, maturation, and rematuration of the species are being studied. The first attempts at hatching S. serrata were successful with rates varying between 75% and 90%. Two out of three trials on larval rearing yielded a few megalops. The first zoeal stages were fed diatoms, rotifers, Artemia salina, and bread yeast. Overfeeding programs were implemented during the critical premolting periods to prevent weakening of the larvae and lessen cannibalism. Larval weakening during the premolt makes them susceptible to attacks by fungi like Lagenidium and ciliates like Vorticella. S. serrata larvae survived salinity levels as low as 15 ppt until the 14th day of rearing. Other larvae were able to survive in salinities of 30-32 ppt for 8 to 13 days. Zoeal molting was hastened by lowering the salinity to 25-27 ppt. Artificial broodstocking of juveniles and adult crabs has been made possible using a simple refuge system made of three-compartmented hollow blocks. This system has been helpful in minimizing fighting among crabs. Remarkable growth rates have been observed with feeds like mussel meat and trash fish. Average growth increments of 11 mm carapace length and 20 . 35 g body weight have been observed every fortnight. A newly spent spawner could gain additional weight of 22 . 5 g in only 6 days. Feeding rates of juveniles and adult crabs have been established based on the average body weight from an experiment using mussel meat. Crabs feed more at night. In another experiment, eyestalk ablation was found to be effective in inducing growth and mating. Aside from hastening the molting process, copulation is induced even among the small crabs (average carapace length = 55 mm). Natural mating lasts about 26 hr. A copulation which lasted for seven days with a break in between was observed.