Automatic speaker recognition under adverse conditions

Speaker verification is the process of verifying the identity of a person by analysing their speech. There are several important applications for automatic speaker verification (ASV) technology including suspect identification, tracking terrorists and detecting a person’s presence at a remote location in the surveillance domain, as well as person authentication for phone banking and credit card transactions in the private sector. Telephones and telephony networks provide a natural medium for these applications. The aim of this work is to improve the usefulness of ASV technology for practical applications in the presence of adverse conditions. In a telephony environment, background noise, handset mismatch, channel distortions, room acoustics and restrictions on the available testing and training data are common sources of errors for ASV systems. Two research themes were pursued to overcome these adverse conditions: Modelling mismatch and modelling uncertainty. To directly address the performance degradation incurred through mismatched conditions it was proposed to directly model this mismatch. Feature mapping was evaluated for combating handset mismatch and was extended through the use of a blind clustering algorithm to remove the need for accurate handset labels for the training data. Mismatch modelling was then generalised by explicitly modelling the session conditions as a constrained offset of the speaker model means. This session variability modelling approach enabled the modelling of arbitrary sources of mismatch, including handset type, and halved the error rates in many cases. Methods to model the uncertainty in speaker model estimates and verification scores were developed to address the difficulties of limited training and testing data. The Bayes factor was introduced to account for the uncertainty of the speaker model estimates in testing by applying Bayesian theory to the verification criterion, with improved performance in matched conditions. Modelling the uncertainty in the verification score itself met with significant success. Estimating a confidence interval for the "true" verification score enabled an order of magnitude reduction in the average quantity of speech required to make a confident verification decision based on a threshold. The confidence measures developed in this work may also have significant applications for forensic speaker verification tasks.

The use of speech and lip modalities for robust speaker verification under adverse conditions

Investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. We have previously shown (Int. Conf. on Acoustics, Speech and Signal Proc., vol. 6, pp. 3693-3696, May 1998) that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms either subsystem individually. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

Selective combination of visual and thermal imaging for resilient localization in adverse conditions : day and night, smoke and fire

Long-term autonomy in robotics requires perception systems that are resilient to unusual but realistic conditions that will eventually occur during extended missions. For example, unmanned ground vehicles (UGVs) need to be capable of operating safely in adverse and low-visibility conditions, such as at night or in the presence of smoke. The key to a resilient UGV perception system lies in the use of multiple sensor modalities, e.g., operating at different frequencies of the electromagnetic spectrum, to compensate for the limitations of a single sensor type. In this paper, visual and infrared imaging are combined in a Visual-SLAM algorithm to achieve localization. We propose to evaluate the quality of data provided by each sensor modality prior to data combination. This evaluation is used to discard low-quality data, i.e., data most likely to induce large localization errors. In this way, perceptual failures are anticipated and mitigated. An extensive experimental evaluation is conducted on data sets collected with a UGV in a range of environments and adverse conditions, including the presence of smoke (obstructing the visual camera), fire, extreme heat (saturating the infrared camera), low-light conditions (dusk), and at night with sudden variations of artificial light. A total of 240 trajectory estimates are obtained using five different variations of data sources and data combination strategies in the localization method. In particular, the proposed approach for selective data combination is compared to methods using a single sensor type or combining both modalities without preselection. We show that the proposed framework allows for camera-based localization resilient to a large range of low-visibility conditions.

Improving the Robustness of Distributed Failure Detectors in Adverse Conditions

Failure detection is at the core of most fault tolerance strategies, but it often depends on reliable communication. We present new algorithms for failure detectors which are appropriate as components of a fault tolerance system that can be deployed in situations of adverse network conditions (such as loosely connected and administered computing grids). It packs redundancy into heartbeat messages, thereby improving on the robustness of the traditional protocols. Results from experimental tests conducted in a simulated environment with adverse network conditions show significant improvement over existing solutions.

Learning object dynamics for smooth tracking of moving lip contours

A new technique is proposed for learning the dynamic characteristics of a deformable object, applied in particular to the problem of lip-tracking. Experimental results are given which demonstrate that the use of dynamic models allows the system to track more robustly under adverse conditions and to correct spurious, poorly tracked frames

Bi-objective bundle adjustment: towards large-scale visual sea-floor mapping with a minimal sensor suite

Visual sea-floor mapping is a rapidly growing application for Autonomous Underwater Vehicles (AUVs). AUVs are well-suited to the task as they remove humans from a potentially dangerous environment, can reach depths human divers cannot, and are capable of long-term operation in adverse conditions. The output of sea-floor maps generated by AUVs has a number of applications in scientific monitoring: from classifying coral in high biological value sites to surveying sea sponges to evaluate marine environment health.

Evaluating temperate species for the subtropics. 1. Annual ryegrasses

In the subtropics of Australia, irrigated temperate species are the key to reliable cool season feed on dairy farms. Persistence of perennial species is a major limitation to achieving reliable production from irrigated areas and yearly sowings of annual ryegrasses have replaced them as the most productive cool season forage production system in the subtropics. This series of experiments evaluated the yield, and resistance to rust damage, of commercially available cultivars and breeders' lines of annually sown ryegrasses (Lolium multiflorum, L. rigidum, L. x boucheanum and L perenne) in pure, nitrogen-fertilised swards under irrigation in the subtropics over a 22-year period. Barberia and Aristocrat 2 were the most adapted cultivars for subtropical conditions, producing high yields (119 and 114% of mean yield, respectively) and demonstrating the least rust damage. Newer selections from New Zealand, South African, United States of America and European breeding programs are performing better under subtropical conditions than older cultivars, particularly if a component of the selection process has been conducted in that environment. Cultivars such as Passerei Plus, Crusader, Hulk, Status and Warrior are examples of this process, producing between 105 and 115% of mean yield. Yields of annual ryegrass cultivars, which have been available or still are available for sale in Australia, ranged from 14-30 t/ha DM, depending on cultivar, site and seasonal conditions. Yields were lower at the site, which had inferior soil structure and drainage. Up to 50% of yield was produced in the 3 winter months. There was a trend towards improved yields and better tolerance of crown rust from experimental lines in the subtropics, as breeders strive for wider adaptation. Around 70% of the variation in total yield of annual ryegrass and 50 and 60% of the variation in winter and spring yield, respectively, were significantly explained by cultivar, site and climatic variables in autumn, winter and spring. While level of rust damage had no effect on total or seasonal yields, it affected the amount of green leaf available in spring. Under subtropical conditions, winter, spring and overall (autumn to mid-summer) temperatures influenced the- development of rust, which along with cultivar, accounted for 46% of the variation in rust damage. Cultivars showed a range of adaptation, with some performing well only under adverse conditions, some being well adapted to all conditions and some which performed well only under favoured conditions. Cultivars with high winter yields were most suited to subtropical conditions and included Aristocrat 2 (now released as CM 108), Barberia, Warrior, Crusader, Status, Passerei Plus and Hulk. Short growing season types such as Winter Star and T Rex performed well in winter but achieved lower total production, and long season cultivars such as Flanker rarely achieved their potential because of unfavourable conditions in late summer.

Continued delivery of applied solutions to weed issues in central Queensland

Developing best practices in Central Queensland to (a) manage difficult to control weeds; (b) improve herbicide efficacy under adverse conditions, and (c) manage weeds in wide-row crop systems.

Effect of substrate roughness on growth of diamond by hot filament CVD

Polycrystalline diamond coatings are grown on Si (100) substrate by hot filament CVD technique. We investigate here the effect of substrate roughening on the substrate temperature and methane concentration required to maintain high quality, high growth rate and faceted morphology of the diamond coatings. It has been shown that as we increase the substrate roughness from 0.05 mu m to 0.91 mu m (centre line average or CLA) there is enhancement in deposited film quality (Raman peak intensity ratio of sp (3) to non-sp (3) content increases from 1.65 to 7.13) and the substrate temperature can be brought down to 640A degrees C without any additional substrate heating. The coatings grown at adverse conditions for sp (3) deposition has cauliflower morphology with nanocrystalline grains and coatings grown under favourable sp (3) condition gives clear faceted grains.

Enhanced Survival of Probiotic Lactobacillus acidophilus by Encapsulation with Nanostructured Polyelectrolyte Layers through Layer-by-Layer Approach

The encapsulation of probiotic Lactobacillus acidophilus through layer-by-layer self-assembly of polyelectrolytes (PE) chitosan (CHI) and carboxymethyl cellulose (CMC) has been investigated,to enhance its survival m adverse conditions encountered in the GI tract The survival of encapsulated cells in simulated gastric (SGF) and intestinal fluids (SIF) is significant when compared to nonencapsulated cells On sequential exposure to SGF and SIF fox 120 nun, almost complete death of free cells is observed However, for cells coated with three nanolayers of PEs (CHI/CMC/CHI) about 33 log % of the cells (6 log cfu/500 mg) survived under the same conditions The enhanced survival rate of encapsulated L acidophilus can be attributed to the impermeability of polyelectrolyte nanolayers to large enzyme molecules like pepsin, and pancreatin that cause proteolysis and to the stability of the polyelectrolyte nanolayers in gastric and intestinal pH The PE coating also serves to reduce viability losses during freezing and freeze- drying About 73 and 92 log % of uncoated and coated cells survived after freeze:drying, and the losses occurring between freezing and freeze-drying were found to be lower for coated cells

HUMAN DETECTION USING SPARSE REPRESENTATION

The problem of human detection is challenging, more so, when faced with adverse conditions such as occlusion and background clutter. This paper addresses the problem of human detection by representing an extracted feature of an image using a sparse linear combination of chosen dictionary atoms. The detection along with the scale finding, is done by using the coefficients obtained from sparse representation. This is of particular interest as we address the problem of scale using a scale-embedded dictionary where the conventional methods detect the object by running the detection window at all scales.

Regulation of transcription initiation in mycobacteria

The success of Mycobacterium tuberculosis as a deadly pathogen lies in its ability to survive under adverse conditions during pre- and post-infectious stages. The transcription process and the regulation of gene expression are central to the survival of the pathogen through the harsh conditions. Multiple sigma factors, transcription regulators, diverse two-component systems contribute in tailoring the events to meet the challenges faced by the pathogen. Although the machinery is conserved, many aspects of transcription and its regulation seem to be different in mycobacteria when compared to the other well-studied organisms. Here, we discuss salient aspects of transcription and its regulation in the context of distinct physiology of mycobacteria.

Short-Term Nonlinear Response of Tension Leg Platform in Random Sea Waves

Most of the existing mathematical models for analyzing the dynamic response of TLP are based on explicit or implicit assumptions that motions (translations and rotations) are small magnitude. However, when TLP works in severe adverse conditions, the a priori assumption on small displacements may be inadequate. In such situation, the motions should be regarded as finite magnitude. This paper will study stochastic nonlinear dynamic responses of TLP with finite displacements in random waves. The nonlinearities considered are: large amplitude motions, coupling the six degrees-of-freedom, instantaneous position, instantaneous wet surface, free surface effects and viscous drag force. The nonlinear dynamic responses are calculated by using numerical integration procedure in the time domain. After the time histories of the dynamic responses are obtained, we carry out cycle counting of the stress histories of the tethers with rain-flow counting method to get the stress range distribution.

Year class strengths of perch and pike in Windermere

The pike (Esox lucius) year classes are more stable than those of the perch (Perca fluviatilis), and have been shown to be closely correlated with temp conditions during the first few months of life. The perch year class strengths have been more variable; for success they require the presence of several positive conditions and the absence of many adverse conditions which could cause failure, a favourable combination of circumstances rarely occurs. The conclusions refer only to Windermere from 1941-1964.