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.

Automatic spoken language identification utilizing acoustic and phonetic speech information

Automatic spoken Language Identi¯cation (LID) is the process of identifying the language spoken within an utterance. The challenge that this task presents is that no prior information is available indicating the content of the utterance or the identity of the speaker. The trend of globalization and the pervasive popularity of the Internet will amplify the need for the capabilities spoken language identi¯ca- tion systems provide. A prominent application arises in call centers dealing with speakers speaking di®erent languages. Another important application is to index or search huge speech data archives and corpora that contain multiple languages. The aim of this research is to develop techniques targeted at producing a fast and more accurate automatic spoken LID system compared to the previous National Institute of Standards and Technology (NIST) Language Recognition Evaluation. Acoustic and phonetic speech information are targeted as the most suitable fea- tures for representing the characteristics of a language. To model the acoustic speech features a Gaussian Mixture Model based approach is employed. Pho- netic speech information is extracted using existing speech recognition technol- ogy. Various techniques to improve LID accuracy are also studied. One approach examined is the employment of Vocal Tract Length Normalization to reduce the speech variation caused by di®erent speakers. A linear data fusion technique is adopted to combine the various aspects of information extracted from speech. As a result of this research, a LID system was implemented and presented for evaluation in the 2003 Language Recognition Evaluation conducted by the NIST.

Real-time performance modelling of a sustained attention to response task

Vigilance declines when exposed to highly predictable and uneventful tasks. Monotonous tasks provide little cognitive and motor stimulation and contribute to human errors. This paper aims to model and detect vigilance decline in real time through participant’s reaction times during a monotonous task. A lab-based experiment adapting the Sustained Attention to Response Task (SART) is conducted to quantify the effect of monotony on overall performance. Then relevant parameters are used to build a model detecting hypovigilance throughout the experiment. The accuracy of different mathematical models are compared to detect in real-time – minute by minute - the lapses in vigilance during the task. We show that monotonous tasks can lead to an average decline in performance of 45%. Furthermore, vigilance modelling enables to detect vigilance decline through reaction times with an accuracy of 72% and a 29% false alarm rate. Bayesian models are identified as a better model to detect lapses in vigilance as compared to Neural Networks and Generalised Linear Mixed Models. This modelling could be used as a framework to detect vigilance decline of any human performing monotonous tasks.

Probabilistic visual recognition of artificial landmarks for simultaneous localization and mapping

Probabilistic robotics, most often applied to the problem of simultaneous localisation and mapping (SLAM), requires measures of uncertainly to accompany observations of the environment. This paper describes how uncertainly can be characterised for a vision system that locates coloured landmark in a typical laboratory environment. The paper describes a model of the uncertainly in segmentation, the internal camera model and the mounting of the camera on the robot. It =plains the implementation of the system on a laboratory robot, and provides experimental results that show the coherence of the uncertainly model,

Wide-baseline keypoint detection and matching with wide-angle images for vision based localisation

This thesis addresses the problem of detecting and describing the same scene points in different wide-angle images taken by the same camera at different viewpoints. This is a core competency of many vision-based localisation tasks including visual odometry and visual place recognition. Wide-angle cameras have a large field of view that can exceed a full hemisphere, and the images they produce contain severe radial distortion. When compared to traditional narrow field of view perspective cameras, more accurate estimates of camera egomotion can be found using the images obtained with wide-angle cameras. The ability to accurately estimate camera egomotion is a fundamental primitive of visual odometry, and this is one of the reasons for the increased popularity in the use of wide-angle cameras for this task. Their large field of view also enables them to capture images of the same regions in a scene taken at very different viewpoints, and this makes them suited for visual place recognition. However, the ability to estimate the camera egomotion and recognise the same scene in two different images is dependent on the ability to reliably detect and describe the same scene points, or ‘keypoints’, in the images. Most algorithms used for this purpose are designed almost exclusively for perspective images. Applying algorithms designed for perspective images directly to wide-angle images is problematic as no account is made for the image distortion. The primary contribution of this thesis is the development of two novel keypoint detectors, and a method of keypoint description, designed for wide-angle images. Both reformulate the Scale- Invariant Feature Transform (SIFT) as an image processing operation on the sphere. As the image captured by any central projection wide-angle camera can be mapped to the sphere, applying these variants to an image on the sphere enables keypoints to be detected in a manner that is invariant to image distortion. Each of the variants is required to find the scale-space representation of an image on the sphere, and they differ in the approaches they used to do this. Extensive experiments using real and synthetically generated wide-angle images are used to validate the two new keypoint detectors and the method of keypoint description. The best of these two new keypoint detectors is applied to vision based localisation tasks including visual odometry and visual place recognition using outdoor wide-angle image sequences. As part of this work, the effect of keypoint coordinate selection on the accuracy of egomotion estimates using the Direct Linear Transform (DLT) is investigated, and a simple weighting scheme is proposed which attempts to account for the uncertainty of keypoint positions during detection. A word reliability metric is also developed for use within a visual ‘bag of words’ approach to place recognition.

An investigation of dose calculation accuracy in intensity-modulated radiotherapy of sites in the head & neck

Knowledge of the accuracy of dose calculations in intensity-modulated radiotherapy of the head and neck is essential for clinical confidence in these highly conformal treatments. High dose gradients are frequently placed very close to critical structures, such as the spinal cord, and good coverage of complex shaped nodal target volumes is important for long term-local control. A phantom study is presented comparing the performance of standard clinical pencil-beam and collapsed-cone dose algorithms to Monte Carlo calculation and three-dimensional gel dosimetry measurement. All calculations and measurements are normalized to the median dose in the primary planning target volume, making this a purely relative study. The phantom simulates tissue, air and bone for a typical neck section and is treated using an inverse-planned 5-field IMRT treatment, similar in character to clinically used class solutions. Results indicate that the pencil-beam algorithm fails to correctly model the relative dose distribution surrounding the air cavity, leading to an overestimate of the target coverage. The collapsed-cone and Monte Carlo results are very similar, indicating that the clinical collapsed-cone algorithm is perfectly sufficient for routine clinical use. The gel measurement shows generally good agreement with the collapsed-cone and Monte Carlo calculated dose, particularly in the spinal cord dose and nodal target coverage, thus giving greater confidence in the use of this class solution.

Focus-score weighted super-resolution for uncooperative iris recognition at a distance and on the move

Uncooperative iris identification systems at a distance and on the move often suffer from poor resolution and poor focus of the captured iris images. The lack of pixel resolution and well-focused images significantly degrades the iris recognition performance. This paper proposes a new approach to incorporate the focus score into a reconstruction-based super-resolution process to generate a high resolution iris image from a low resolution and focus inconsistent video sequence of an eye. A reconstruction-based technique, which can incorporate middle and high frequency components from multiple low resolution frames into one desired super-resolved frame without introducing false high frequency components, is used. A new focus assessment approach is proposed for uncooperative iris at a distance and on the move to improve performance for variations in lighting, size and occlusion. A novel fusion scheme is then proposed to incorporate the proposed focus score into the super-resolution process. The experiments conducted on the The Multiple Biometric Grand Challenge portal database shows that our proposed approach achieves an EER of 2.1%, outperforming the existing state-of-the-art averaging signal-level fusion approach by 19.2% and the robust mean super-resolution approach by 8.7%.

Continuous biometric authentication : can it be more practical?

Continuous biometric authentication schemes (CBAS) are built around the biometrics supplied by user behavioural characteristics and continuously check the identity of the user throughout the session. The current literature for CBAS primarily focuses on the accuracy of the system in order to reduce false alarms. However, these attempts do not consider various issues that might affect practicality in real world applications and continuous authentication scenarios. One of the main issues is that the presented CBAS are based on several samples of training data either of both intruder and valid users or only the valid users' profile. This means that historical profiles for either the legitimate users or possible attackers should be available or collected before prediction time. However, in some cases it is impractical to gain the biometric data of the user in advance (before detection time). Another issue is the variability of the behaviour of the user between the registered profile obtained during enrollment, and the profile from the testing phase. The aim of this paper is to identify the limitations in current CBAS in order to make them more practical for real world applications. Also, the paper discusses a new application for CBAS not requiring any training data either from intruders or from valid users.

Low-cost hardware speech enhancement for improved speech recognition in automotive environments

Voice recognition is one of the key enablers to reduce driver distraction as in-vehicle systems become more and more complex. With the integration of voice recognition in vehicles, safety and usability are improved as the driver’s eyes and hands are not required to operate system controls. Whilst speaker independent voice recognition is well developed, performance in high noise environments (e.g. vehicles) is still limited. La Trobe University and Queensland University of Technology have developed a low-cost hardware-based speech enhancement system for automotive environments based on spectral subtraction and delay–sum beamforming techniques. The enhancement algorithms have been optimised using authentic Australian English collected under typical driving conditions. Performance tests conducted using speech data collected under variety of vehicle noise conditions demonstrate a word recognition rate improvement in the order of 10% or more under the noisiest conditions. Currently developed to a proof of concept stage there is potential for even greater performance improvement.