The QUT-NOISE-SRE protocol for the evaluation of noisy speaker recognition

The QUT-NOISE-SRE protocol is designed to mix the large QUT-NOISE database, consisting of over 10 hours of back- ground noise, collected across 10 unique locations covering 5 common noise scenarios, with commonly used speaker recognition datasets such as Switchboard, Mixer and the speaker recognition evaluation (SRE) datasets provided by NIST. By allowing common, clean, speech corpora to be mixed with a wide variety of noise conditions, environmental reverberant responses, and signal-to-noise ratios, this protocol provides a solid basis for the development, evaluation and benchmarking of robust speaker recognition algorithms, and is freely available to download alongside the QUT-NOISE database. In this work, we use the QUT-NOISE-SRE protocol to evaluate a state-of-the-art PLDA i-vector speaker recognition system, demonstrating the importance of designing voice-activity-detection front-ends specifically for speaker recognition, rather than aiming for perfect coherence with the true speech/non-speech boundaries.

Speaker identification using higher order spectral phase features and their effectiveness vis-a-vis Mel-Cepstral features

The effectiveness of higher-order spectral (HOS) phase features in speaker recognition is investigated by comparison with Mel Cepstral features on the same speech data. HOS phase features retain phase information from the Fourier spectrum unlikeMel–frequency Cepstral coefficients (MFCC). Gaussian mixture models are constructed from Mel– Cepstral features and HOS features, respectively, for the same data from various speakers in the Switchboard telephone Speech Corpus. Feature clusters, model parameters and classification performance are analyzed. HOS phase features on their own provide a correct identification rate of about 97% on the chosen subset of the corpus. This is the same level of accuracy as provided by MFCCs. Cluster plots and model parameters are compared to show that HOS phase features can provide complementary information to better discriminate between speakers.

Improved GMM-based speaker verification using SVM-driven impostor dataset selection

The problem of impostor dataset selection for GMM-based speaker verification is addressed through the recently proposed data-driven background dataset refinement technique. The SVM-based refinement technique selects from a candidate impostor dataset those examples that are most frequently selected as support vectors when training a set of SVMs on a development corpus. This study demonstrates the versatility of dataset refinement in the task of selecting suitable impostor datasets for use in GMM-based speaker verification. The use of refined Z- and T-norm datasets provided performance gains of 15% in EER in the NIST 2006 SRE over the use of heuristically selected datasets. The refined datasets were shown to generalise well to the unseen data of the NIST 2008 SRE.

Data-driven impostor selection for T-norm score normalisation and the background dataset in SVM-based speaker verification

A data-driven background dataset refinement technique was recently proposed for SVM based speaker verification. This method selects a refined SVM background dataset from a set of candidate impostor examples after individually ranking examples by their relevance. This paper extends this technique to the refinement of the T-norm dataset for SVM-based speaker verification. The independent refinement of the background and T-norm datasets provides a means of investigating the sensitivity of SVM-based speaker verification performance to the selection of each of these datasets. Using refined datasets provided improvements of 13% in min. DCF and 9% in EER over the full set of impostor examples on the 2006 SRE corpus with the majority of these gains due to refinement of the T-norm dataset. Similar trends were observed for the unseen data of the NIST 2008 SRE.

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.

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.

Spectral coding methods for speech compression and speaker identification

This thesis investigates aspects of encoding the speech spectrum at low bit rates, with extensions to the effect of such coding on automatic speaker identification. Vector quantization (VQ) is a technique for jointly quantizing a block of samples at once, in order to reduce the bit rate of a coding system. The major drawback in using VQ is the complexity of the encoder. Recent research has indicated the potential applicability of the VQ method to speech when product code vector quantization (PCVQ) techniques are utilized. The focus of this research is the efficient representation, calculation and utilization of the speech model as stored in the PCVQ codebook. In this thesis, several VQ approaches are evaluated, and the efficacy of two training algorithms is compared experimentally. It is then shown that these productcode vector quantization algorithms may be augmented with lossless compression algorithms, thus yielding an improved overall compression rate. An approach using a statistical model for the vector codebook indices for subsequent lossless compression is introduced. This coupling of lossy compression and lossless compression enables further compression gain. It is demonstrated that this approach is able to reduce the bit rate requirement from the current 24 bits per 20 millisecond frame to below 20, using a standard spectral distortion metric for comparison. Several fast-search VQ methods for use in speech spectrum coding have been evaluated. The usefulness of fast-search algorithms is highly dependent upon the source characteristics and, although previous research has been undertaken for coding of images using VQ codebooks trained with the source samples directly, the product-code structured codebooks for speech spectrum quantization place new constraints on the search methodology. The second major focus of the research is an investigation of the effect of lowrate spectral compression methods on the task of automatic speaker identification. The motivation for this aspect of the research arose from a need to simultaneously preserve the speech quality and intelligibility and to provide for machine-based automatic speaker recognition using the compressed speech. This is important because there are several emerging applications of speaker identification where compressed speech is involved. Examples include mobile communications where the speech has been highly compressed, or where a database of speech material has been assembled and stored in compressed form. Although these two application areas have the same objective - that of maximizing the identification rate - the starting points are quite different. On the one hand, the speech material used for training the identification algorithm may or may not be available in compressed form. On the other hand, the new test material on which identification is to be based may only be available in compressed form. Using the spectral parameters which have been stored in compressed form, two main classes of speaker identification algorithm are examined. Some studies have been conducted in the past on bandwidth-limited speaker identification, but the use of short-term spectral compression deserves separate investigation. Combining the major aspects of the research, some important design guidelines for the construction of an identification model when based on the use of compressed speech are put forward.

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.

Vector quantization based Gaussian modeling for speaker verification

Gaussian mixture models (GMMs) have become an established means of modeling feature distributions in speaker recognition systems. It is useful for experimentation and practical implementation purposes to develop and test these models in an efficient manner particularly when computational resources are limited. A method of combining vector quantization (VQ) with single multi-dimensional Gaussians is proposed to rapidly generate a robust model approximation to the Gaussian mixture model. A fast method of testing these systems is also proposed and implemented. Results on the NIST 1996 Speaker Recognition Database suggest comparable and in some cases an improved verification performance to the traditional GMM based analysis scheme. In addition, previous research for the task of speaker identification indicated a similar system perfomance between the VQ Gaussian based technique and GMMs