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.

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.

Laser-to-radar sensing redundancy for resilient perception in adverse environmental conditions

This paper presents an approach to promote the integrity of perception systems for outdoor unmanned ground vehicles (UGV) operating in challenging environmental conditions (presence of dust or smoke). The proposed technique automatically evaluates the consistency of the data provided by two sensing modalities: a 2D laser range finder and a millimetre-wave radar, allowing for perceptual failure mitigation. Experimental results, obtained with a UGV operating in rural environments, and an error analysis validate the approach.

Adherence to medicines in the older-aged with chronic conditions : does an intervention concerning adherence by an allied health professional help?

Adherence to medicines is a major determinant of the effectiveness of medicines. However, estimates of non-adherence in the older-aged with chronic conditions vary from 40 to 75%. The problems caused by non-adherence in the older-aged include residential care and hospital admissions, progression of the disease, and increased costs to society. The reasons for non-adherence in the older-aged include items related to the medicine (e.g. cost, number of medicines, adverse effects) and those related to person (e.g. cognition, vision, depression). It is also known that there are many ways adherence can be increased (e.g. use of blister packs, cues). It is assumed that interventions by allied health professions, including a discussion of adherence, will improve adherence to medicines in the older aged but the evidence for this has not been reviewed. There is some evidence that telephone counselling about adherence by a nurse or pharmacist does improve adherence, short- and long-term. However, face-to-face intervention counselling at the pharmacy, or during a home visit by a pharmacist, has shown variable results with some studies showing improved adherence and some not. Education programs during hospital stays have not been shown to improve adherence on discharge, but education programs for subjects with hypertension have been shown to improve adherence. In combination with an education program, both counselling and a medicine review program have been shown to improve adherence short-term in the older-aged. Thus, there are many unanswered questions about the most effective interventions to promote adherence. More studies are needed to determine the most appropriate interventions by allied health professions, and these need to consider the disease state, demographics, and socio-economic status of the older-aged subject, and the intensity and duration of intervention needed.

Towards reliable perception for unmanned ground vehicles in challenging conditions

This work aims to promote reliability and integrity in autonomous perceptual systems, with a focus on outdoor unmanned ground vehicle (UGV) autonomy. For this purpose, a comprehensive UGV system, comprising many different exteroceptive and proprioceptive sensors has been built. The first contribution of this work is a large, accurately calibrated and synchronised, multi-modal data-set, gathered in controlled environmental conditions, including the presence of dust, smoke and rain. The data have then been used to analyse the effects of such challenging conditions on perception and to identify common perceptual failures. The second contribution is a presentation of methods for mitigating these failures to promote perceptual integrity in adverse environmental conditions.

Adverse weather effects on bus ridership

This study focuses on weather effects on daily bus ridership in Brisbane, given bus’ dominance in this city. The weather pattern of Brisbane varies by season according to its sub-tropical climate characteristics. Bus is prone to inclement weather condition as it shares the road system with general traffic. Moreover, bus stops generally offer less or sometimes no protection from adverse weather. Hence, adverse weather conditions such as rain are conjectured to directly impact on daily travel behaviour patterns. There has been limited Australian research on the impact of weather on daily transit ridership. This study investigates the relationship between rainy day and daily bus ridership for the period of 2010 to 2012. Overall, rainfall affects negatively with varying impacts on different transit groups. However, this analysis confirmed a positive relationship between consecutive rainy days (rain continuing for 3 or more days). A possible explanation could be that people may switch their transport mode to bus to avoid high traffic congestion and higher accident potentiality on rainy days. Also, Brisbane’s segregated busway (BRT) corridor works favourably towards this mode choice. Our study findings enhance the fundamental understanding of traveller behaviour, particularly mode choice behaviour under adverse weather conditions.

Marginal costs of hospital-acquired conditions : information for priority-setting for patient safety programmes and research

Objective: To estimate the relative inpatient costs of hospital-acquired conditions. Methods: Patient level costs were estimated using computerized costing systems that log individual utilization of inpatient services and apply sophisticated cost estimates from the hospital's general ledger. Occurrence of hospital-acquired conditions was identified using an Australian ‘condition-onset' flag for diagnoses not present on admission. These were grouped to yield a comprehensive set of 144 categories of hospital-acquired conditions to summarize data coded with ICD-10. Standard linear regression techniques were used to identify the independent contribution of hospital-acquired conditions to costs, taking into account the case-mix of a sample of acute inpatients (n = 1,699,997) treated in Australian public hospitals in Victoria (2005/06) and Queensland (2006/07). Results: The most costly types of complications were post-procedure endocrine/metabolic disorders, adding AU$21,827 to the cost of an episode, followed by MRSA (AU$19,881) and enterocolitis due to Clostridium difficile (AU$19,743). Aggregate costs to the system, however, were highest for septicaemia (AU$41.4 million), complications of cardiac and vascular implants other than septicaemia (AU$28.7 million), acute lower respiratory infections, including influenza and pneumonia (AU$27.8 million) and UTI (AU$24.7 million). Hospital-acquired complications are estimated to add 17.3% to treatment costs in this sample. Conclusions: Patient safety efforts frequently focus on dramatic but rare complications with very serious patient harm. Previous studies of the costs of adverse events have provided information on ‘indicators’ of safety problems rather than the full range of hospital-acquired conditions. Adding a cost dimension to priority-setting could result in changes to the focus of patient safety programmes and research. Financial information should be combined with information on patient outcomes to allow for cost-utility evaluation of future interventions.

A process mining-based investigation of adverse events in care processes

This paper proposes the Clinical Pathway Analysis Method (CPAM) approach that enables the extraction of valuable organisational and medical information on past clinical pathway executions from the event logs of healthcare information systems. The method deals with the complexity of real-world clinical pathways by introducing a perspective-based segmentation of the date-stamped event log. CPAM enables the clinical pathway analyst to effectively and efficiently acquire a profound insight into the clinical pathways. By comparing the specific medical conditions of patients with the factors used for characterising the different clinical pathway variants, the medical expert can identify the best therapeutic option. Process mining-based analytics enables the acquisition of valuable insights into clinical pathways, based on the complete audit traces of previous clinical pathway instances. Additionally, the methodology is suited to assess guideline compliance and analyse adverse events. Finally, the methodology provides support for eliciting tacit knowledge and providing treatment selection assistance.

Adverse weather effects on bus ridership

This study focuses on the effects of weather on daily bus ridership in Brisbane, given the dominance of buses in that city. The weather pattern of Brisbane varies by season according to its subtropical climate characteristics. Bus operation is affected by inclement weather conditions, as buses share the road system with general traffic. Moreover, bus stops generally offer little, or sometimes no, protection from adverse weather. Hence, adverse weather conditions such as rain are thought to directly impact on daily travel behaviour patterns. There has been limited Australian research on the impact of weather on daily transit ridership. This study investigates the relationship between rainy days and daily bus ridership for the period 2010 to 2012. Overall, rainfall has a negative effect, with varying impacts on different transit groups. However, this analysis confirmed a positive relationship between consecutive rainy days (rain continuing for 3 or more days). A possible explanation could be that people switch their transport mode to bus to avoid high traffic congestion and higher accident potentiality on rainy days. Also, Brisbane’s segregated busway corridor works favourably towards this mode choice. The findings of our study enhance the fundamental understanding of traveller behaviour, particularly mode-choice behaviour, under adverse weather conditions.