Cascading appearance-based features for visual speaker verification

The cascading appearance-based (CAB) feature extraction technique has established itself as the state of the art in extracting dynamic visual speech features for speech recognition. In this paper, we will focus on investigating the effectiveness of this technique for the related speaker verification application. By investigating the speaker verification ability of each stage of the cascade we will demonstrate that the same steps taken to reduce static speaker and environmental information for the speech recognition application also provide similar improvements for speaker recognition. These results suggest that visual speaker recognition can improve considerable when conducted solely through a consideration of the dynamic speech information rather than the static appearance of the speaker's mouth region.

Correctness ensuring process configuration : an approach based on partner synthesis

A configurable process model describes a family of similar process models in a given domain. Such a model can be configured to obtain a specific process model that is subsequently used to handle individual cases, for instance, to process customer orders. Process configuration is notoriously difficult as there may be all kinds of interdependencies between configuration decisions.} In fact, an incorrect configuration may lead to behavioral issues such as deadlocks and livelocks. To address this problem, we present a novel verification approach inspired by the ``operating guidelines'' used for partner synthesis. We view the configuration process as an external service, and compute a characterization of all such services which meet particular requirements using the notion of configuration guideline. As a result, we can characterize all feasible configurations (i.\,e., configurations without behavioral problems) at design time, instead of repeatedly checking each individual configuration while configuring a process model.

Improving automatic speaker verification using SVM techniques

Automatic recognition of people is an active field of research with important forensic and security applications. In these applications, it is not always possible for the subject to be in close proximity to the system. Voice represents a human behavioural trait which can be used to recognise people in such situations. Automatic Speaker Verification (ASV) is the process of verifying a persons identity through the analysis of their speech and enables recognition of a subject at a distance over a telephone channel { wired or wireless. A significant amount of research has focussed on the application of Gaussian mixture model (GMM) techniques to speaker verification systems providing state-of-the-art performance. GMM's are a type of generative classifier trained to model the probability distribution of the features used to represent a speaker. Recently introduced to the field of ASV research is the support vector machine (SVM). An SVM is a discriminative classifier requiring examples from both positive and negative classes to train a speaker model. The SVM is based on margin maximisation whereby a hyperplane attempts to separate classes in a high dimensional space. SVMs applied to the task of speaker verification have shown high potential, particularly when used to complement current GMM-based techniques in hybrid systems. This work aims to improve the performance of ASV systems using novel and innovative SVM-based techniques. Research was divided into three main themes: session variability compensation for SVMs; unsupervised model adaptation; and impostor dataset selection. The first theme investigated the differences between the GMM and SVM domains for the modelling of session variability | an aspect crucial for robust speaker verification. Techniques developed to improve the robustness of GMMbased classification were shown to bring about similar benefits to discriminative SVM classification through their integration in the hybrid GMM mean supervector SVM classifier. Further, the domains for the modelling of session variation were contrasted to find a number of common factors, however, the SVM-domain consistently provided marginally better session variation compensation. Minimal complementary information was found between the techniques due to the similarities in how they achieved their objectives. The second theme saw the proposal of a novel model for the purpose of session variation compensation in ASV systems. Continuous progressive model adaptation attempts to improve speaker models by retraining them after exploiting all encountered test utterances during normal use of the system. The introduction of the weight-based factor analysis model provided significant performance improvements of over 60% in an unsupervised scenario. SVM-based classification was then integrated into the progressive system providing further benefits in performance over the GMM counterpart. Analysis demonstrated that SVMs also hold several beneficial characteristics to the task of unsupervised model adaptation prompting further research in the area. In pursuing the final theme, an innovative background dataset selection technique was developed. This technique selects the most appropriate subset of examples from a large and diverse set of candidate impostor observations for use as the SVM background by exploiting the SVM training process. This selection was performed on a per-observation basis so as to overcome the shortcoming of the traditional heuristic-based approach to dataset selection. Results demonstrate the approach to provide performance improvements over both the use of the complete candidate dataset and the best heuristically-selected dataset whilst being only a fraction of the size. The refined dataset was also shown to generalise well to unseen corpora and be highly applicable to the selection of impostor cohorts required in alternate techniques for speaker verification.

QUT speaker identity verification system for EVALITA 2009

This document outlines the system submitted by the Speech and Audio Research Laboratory at the Queensland University of Technology (QUT) for the Speaker Identity Verication: Application task of EVALITA 2009. This submission consisted of a score-level fusion of three component systems, a joint-factor GMM system and two SVM systems using GLDS and GMM supervector kernels. Development and evaluation results are presented, demonstrating the effectiveness of this fused system approach.

Data-driven background dataset selection for SVM-based speaker verification

The recently proposed data-driven background dataset refinement technique provides a means of selecting an informative background for support vector machine (SVM)-based speaker verification systems. This paper investigates the characteristics of the impostor examples in such highly-informative background datasets. Data-driven dataset refinement individually evaluates the suitability of candidate impostor examples for the SVM background prior to selecting the highest-ranking examples as a refined background dataset. Further, the characteristics of the refined dataset were analysed to investigate the desired traits of an informative SVM background. The most informative examples of the refined dataset were found to consist of large amounts of active speech and distinctive language characteristics. The data-driven refinement technique was shown to filter the set of candidate impostor examples to produce a more disperse representation of the impostor population in the SVM kernel space, thereby reducing the number of redundant and less-informative examples in the background dataset. Furthermore, data-driven refinement was shown to provide performance gains when applied to the difficult task of refining a small candidate dataset that was mis-matched to the evaluation conditions.

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.

Dynamic visual features for audio-visual speaker verification

The cascading appearance-based (CAB) feature extraction technique has established itself as the state-of-the-art in extracting dynamic visual speech features for speech recognition. In this paper, we will focus on investigating the effectiveness of this technique for the related speaker verification application. By investigating the speaker verification ability of each stage of the cascade we will demonstrate that the same steps taken to reduce static speaker and environmental information for the visual speech recognition application also provide similar improvements for visual speaker recognition. A further study is conducted comparing synchronous HMM (SHMM) based fusion of CAB visual features and traditional perceptual linear predictive (PLP) acoustic features to show that higher complexity inherit in the SHMM approach does not appear to provide any improvement in the final audio-visual speaker verification system over simpler utterance level score fusion.

Universal online interventions might engage psychologically distressed university students who are unlikely to seek formal help

University students are a high risk population for mental health problems, yet few seek professional help when experiencing problems. This study explored the potential role of an online intervention for promoting wellbeing in university students, by investigating students' help-seeking behaviour, intention to use online interventions and student content preference for such interventions; 254 university students responded to an online survey designed for this study. As predicted, students were less likely to seek help as levels of psychological distress increased. Conversely, intention to use an online intervention increased at higher levels of distress, with 39.1%, 49.4% and 57.7% of low, moderate and severely distressed students respectively indicating they would use an online program supporting student well-being. Results suggest that online interventions may be a useful way to provide help to students in need who otherwise may not seek formal help.

Applying SVMs and weight-based factor analysis to unsupervised adaptation for speaker verification

This paper presents an extended study on the implementation of support vector machine(SVM) based speaker verification in systems that employ continuous progressive model adaptation using the weight-based factor analysis model. The weight-based factor analysis model compensates for session variations in unsupervised scenarios by incorporating trial confidence measures in the general statistics used in the inter-session variability modelling process. Employing weight-based factor analysis in Gaussian mixture models (GMM) was recently found to provide significant performance gains to unsupervised classification. Further improvements in performance were found through the integration of SVM-based classification in the system by means of GMM supervectors. This study focuses particularly on the way in which a client is represented in the SVM kernel space using single and multiple target supervectors. Experimental results indicate that training client SVMs using a single target supervector maximises performance while exhibiting a certain robustness to the inclusion of impostor training data in the model. Furthermore, the inclusion of low-scoring target trials in the adaptation process is investigated where they were found to significantly aid performance.

Settlement geography of African refugee communities in Southeast Queensland : an analysis of residential distribution and secondary migration

Before 2001, most Africans immigrating to Australia were white South Africans and Zimbabweans who arrived as economic and family-reunion migrants (Cox, Cooper & Adepoju, 1999). Black African communities are a more recent addition to the Australian landscape, with most entering Australia as refugees after 2001. African refugees are a particularly disadvantaged immigrant group, which the Department of Immigration and Multicultural Affairs (in the Community Relations Commission of New South Wales, 2006) suggests require high levels of settlement support (p.23). Decision makers and settlement service providers need to have settlement data on the communities so that they can be effective in planning, budgeting and delivering support where it is most needed. Settlement data are also useful for determining the challenges that these communities face in trying to establish themselves in resettlement. There has been no verification of existing secondary data sources, however, or previous formal study of African refugee settlement geography in Southeast Queensland. This research addresses the knowledge gap by using a mixed-method approach to identify and describe the distribution and population size of eight African communities in Southeast Queensland, examine secondary migration patterns in these communities and assess the relationship between these geographic features and housing, a critical factor in successful settlement. Significant discrepancies exist between the primary data gathered in the study and existing secondary data relating to population size and distribution of the communities. Results also reveal a tension between the socio-cultural forces and the housing and economic imperatives driving secondary migration in the communities, and a general lack of engagement by African refugees with structured support networks. These findings have a wide range of implications for policy and for groups that provide settlement support to these communities.

When sparks fly : developing formal mentoring programs for the career development of young and emerging artists

Formal mentoring programs are accepted as a valuable strategy for developing young and emerging artists. This thesis presents the results of an evaluation of the SPARK National Young Artists Mentoring Program (SPARK). SPARK was a ten-month formal mentoring program managed by Youth Arts Queensland (YAQ) on behalf of the Australia Council for the Arts from 2003-2009. The program aimed to assist young and emerging Australian artists between the ages of 18-26 to establish a professional career in the arts. It was a highly successful formal arts mentoring program that facilitated 58 mentorships between young and emerging artists and professional artists from across Australia in five program rounds over its seven year lifespan. Interest from other cultural organisations looking to develop their own formal mentoring programs encouraged YAQ to commission this research to determine how the program works to achieve its effects. This study was conducted with young and emerging artists who participated in SPARK from 2003 to 2008. It took a theory-driven evaluation approach to examine SPARK as an example of what makes formal arts mentoring programs effective. It focused on understanding the program’s theory or how the program worked to achieve its desired outcomes. The program activities and assumed responses to program activities were mapped out in a theories of change model. This theoretical framework was then used to plan the points for data collection. Through the process of data collection, actual program developments were compared to the theoretical framework to see what occurred as expected and what did not. The findings were then generalised for knowledge and wider application. The findings demonstrated that SPARK was a successful and effective program and an exemplar model of a formal mentoring program preparing young and emerging artists for professional careers in the arts. They also indicate several ways in which this already strong program could be further improved, including: looking at the way mentoring relationships are set up and how the mentoring process is managed; considering the balance between artistic and professional development; developing career development competencies and networking skills; taking into account the needs of young and emerging artists to develop their professional identity and build confidence; and giving more thought to the desired program outcomes and considering the issue of timeliness and readiness for career transition. From these findings, together with principles outlined in the mentoring and career development literature, a number of necessary conditions have been identified for developing effective mentoring programs in the career development of young and emerging artists.

Strengthening and formally verifying privacy in identity management systems

In a digital world, users’ Personally Identifiable Information (PII) is normally managed with a system called an Identity Management System (IMS). There are many types of IMSs. There are situations when two or more IMSs need to communicate with each other (such as when a service provider needs to obtain some identity information about a user from a trusted identity provider). There could be interoperability issues when communicating parties use different types of IMS. To facilitate interoperability between different IMSs, an Identity Meta System (IMetS) is normally used. An IMetS can, at least theoretically, join various types of IMSs to make them interoperable and give users the illusion that they are interacting with just one IMS. However, due to the complexity of an IMS, attempting to join various types of IMSs is a technically challenging task, let alone assessing how well an IMetS manages to integrate these IMSs. The first contribution of this thesis is the development of a generic IMS model called the Layered Identity Infrastructure Model (LIIM). Using this model, we develop a set of properties that an ideal IMetS should provide. This idealized form is then used as a benchmark to evaluate existing IMetSs. Different types of IMS provide varying levels of privacy protection support. Unfortunately, as observed by Jøsang et al (2007), there is insufficient privacy protection in many of the existing IMSs. In this thesis, we study and extend a type of privacy enhancing technology known as an Anonymous Credential System (ACS). In particular, we extend the ACS which is built on the cryptographic primitives proposed by Camenisch, Lysyanskaya, and Shoup. We call this system the Camenisch, Lysyanskaya, Shoup - Anonymous Credential System (CLS-ACS). The goal of CLS-ACS is to let users be as anonymous as possible. Unfortunately, CLS-ACS has problems, including (1) the concentration of power to a single entity - known as the Anonymity Revocation Manager (ARM) - who, if malicious, can trivially reveal a user’s PII (resulting in an illegal revocation of the user’s anonymity), and (2) poor performance due to the resource-intensive cryptographic operations required. The second and third contributions of this thesis are the proposal of two protocols that reduce the trust dependencies on the ARM during users’ anonymity revocation. Both protocols distribute trust from the ARM to a set of n referees (n > 1), resulting in a significant reduction of the probability of an anonymity revocation being performed illegally. The first protocol, called the User Centric Anonymity Revocation Protocol (UCARP), allows a user’s anonymity to be revoked in a user-centric manner (that is, the user is aware that his/her anonymity is about to be revoked). The second protocol, called the Anonymity Revocation Protocol with Re-encryption (ARPR), allows a user’s anonymity to be revoked by a service provider in an accountable manner (that is, there is a clear mechanism to determine which entity who can eventually learn - and possibly misuse - the identity of the user). The fourth contribution of this thesis is the proposal of a protocol called the Private Information Escrow bound to Multiple Conditions Protocol (PIEMCP). This protocol is designed to address the performance issue of CLS-ACS by applying the CLS-ACS in a federated single sign-on (FSSO) environment. Our analysis shows that PIEMCP can both reduce the amount of expensive modular exponentiation operations required and lower the risk of illegal revocation of users’ anonymity. Finally, the protocols proposed in this thesis are complex and need to be formally evaluated to ensure that their required security properties are satisfied. In this thesis, we use Coloured Petri nets (CPNs) and its corresponding state space analysis techniques. All of the protocols proposed in this thesis have been formally modeled and verified using these formal techniques. Therefore, the fifth contribution of this thesis is a demonstration of the applicability of CPN and its corresponding analysis techniques in modeling and verifying privacy enhancing protocols. To our knowledge, this is the first time that CPN has been comprehensively applied to model and verify privacy enhancing protocols. From our experience, we also propose several CPN modeling approaches, including complex cryptographic primitives (such as zero-knowledge proof protocol) modeling, attack parameterization, and others. The proposed approaches can be applied to other security protocols, not just privacy enhancing protocols.