Within-session variability modelling for factor analysis speaker verification

This work presents an extended Joint Factor Analysis model including explicit modelling of unwanted within-session variability. The goals of the proposed extended JFA model are to improve verification performance with short utterances by compensating for the effects of limited or imbalanced phonetic coverage, and to produce a flexible JFA model that is effective over a wide range of utterance lengths without adjusting model parameters such as retraining session subspaces. Experimental results on the 2006 NIST SRE corpus demonstrate the flexibility of the proposed model by providing competitive results over a wide range of utterance lengths without retraining and also yielding modest improvements in a number of conditions over current state-of-the-art.

Minimising speaker verification utterance length through confidence based early verification decisions

This paper presents a novel approach of estimating the confidence interval of speaker verification scores. This approach is utilised to minimise the utterance lengths required in order to produce a confident verification decision. The confidence estimation method is also extended to address both the problem of high correlation in consecutive frame scores, and robustness with very limited training samples. The proposed technique achieves a drastic reduction in the typical data requirements for producing confident decisions in an automatic speaker verification system. When evaluated on the NIST 2005 SRE, the early verification decision method demonstrates that an average of 5–10 seconds of speech is sufficient to produce verification rates approaching those achieved previously using an average in excess of 100 seconds of speech.

HCX : an efficient hybrid clustering approach for XML documents

This paper proposes a novel Hybrid Clustering approach for XML documents (HCX) that first determines the structural similarity in the form of frequent subtrees and then uses these frequent subtrees to represent the constrained content of the XML documents in order to determine the content similarity. The empirical analysis reveals that the proposed method is scalable and accurate.

XCFS - a novel approach for clustering XML documents using both the structure and the content

XML document clustering is essential for many document handling applications such as information storage, retrieval, integration and transformation. An XML clustering algorithm should process both the structural and the content information of XML documents in order to improve the accuracy and meaning of the clustering solution. However, the inclusion of both kinds of information in the clustering process results in a huge overhead for the underlying clustering algorithm because of the high dimensionality of the data. This paper introduces a novel approach that first determines the structural similarity in the form of frequent subtrees and then uses these frequent subtrees to represent the constrained content of the XML documents in order to determine the content similarity. The proposed method reduces the high dimensionality of input data by using only the structure-constrained content. The empirical analysis reveals that the proposed method can effectively cluster even very large XML datasets and outperform other existing methods.

Unsupervised speaker adaptation for telephone call transcription

The use of the PC and Internet for placing telephone calls will present new opportunities to capture vast amounts of un-transcribed speech for a particular speaker. This paper investigates how to best exploit this data for speaker-dependent speech recognition. Supervised and unsupervised experiments in acoustic model and language model adaptation are presented. Using one hour of automatically transcribed speech per speaker with a word error rate of 36.0%, unsupervised adaptation resulted in an absolute gain of 6.3%, equivalent to 70% of the gain from the supervised case, with additional adaptation data likely to yield further improvements. LM adaptation experiments suggested that although there seems to be a small degree of speaker idiolect, adaptation to the speaker alone, without considering the topic of the conversation, is in itself unlikely to improve transcription accuracy.

Enhancing an Incremental Clustering Algorithm for Web Page Collections

With the size and state of the Internet today, a good quality approach to organizing this mass of information is of great importance. Clustering web pages into groups of similar documents is one approach, but relies heavily on good feature extraction and document representation as well as a good clustering approach and algorithm. Due to the changing nature of the Internet, resulting in a dynamic dataset, an incremental approach is preferred. In this work we propose an enhanced incremental clustering approach to develop a better clustering algorithm that can help to better organize the information available on the Internet in an incremental fashion. Experiments show that the enhanced algorithm outperforms the original histogram based algorithm by up to 7.5%.

Linkage and heritability analysis of migraine symptom groupings: A comparison of three different clustering methods on twin data

Migraine is a painful disorder for which the etiology remains obscure. Diagnosis is largely based on International Headache Society criteria. However, no feature occurs in all patients who meet these criteria, and no single symptom is required for diagnosis. Consequently, this definition may not accurately reflect the phenotypic heterogeneity or genetic basis of the disorder. Such phenotypic uncertainty is typical for complex genetic disorders and has encouraged interest in multivariate statistical methods for classifying disease phenotypes. We applied three popular statistical phenotyping methods—latent class analysis, grade of membership and grade of membership “fuzzy” clustering (Fanny)—to migraine symptom data, and compared heritability and genome-wide linkage results obtained using each approach. Our results demonstrate that different methodologies produce different clustering structures and non-negligible differences in subsequent analyses. We therefore urge caution in the use of any single approach and suggest that multiple phenotyping methods be used.

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.

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.

Efficient clustering approach for intrusion detection in adhoc networks

Mobile ad-hoc networks (MANETs) are temporary wireless networks useful in emergency rescue services, battlefields operations, mobile conferencing and a variety of other useful applications. Due to dynamic nature and lack of centralized monitoring points, these networks are highly vulnerable to attacks. Intrusion detection systems (IDS) provide audit and monitoring capabilities that offer the local security to a node and help to perceive the specific trust level of other nodes. We take benefit of the clustering concept in MANETs for the effective communication between nodes, where each cluster involves a number of member nodes and is managed by a cluster-head. It can be taken as an advantage in these battery and memory constrained networks for the purpose of intrusion detection, by separating tasks for the head and member nodes, at the same time providing opportunity for launching collaborative detection approach. The clustering schemes are generally used for the routing purposes to enhance the route efficiency. However, the effect of change of a cluster tends to change the route; thus degrades the performance. This paper presents a low overhead clustering algorithm for the benefit of detecting intrusion rather than efficient routing. It also discusses the intrusion detection techniques with the help of this simplified clustering scheme.