Thesis (M. Phil.)--University of Hong Kong, 1990.

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2002. / Includes bibliographical references (leaves 92-96). Also available in electronic version. Access restricted to campus users.

Thesis (M.S.)--Rochester Institute of Technology, 1993. / Typescript. Includes bibliographical references (leaves 81-83).

Thesis (M.S.)--University of Missouri-Columbia, 2005. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file viewed on (January 11, 2007) Vita. Includes bibliographical references.

Thesis (Ph.D.)--University of Texas at Dallas, 2007. / Includes vita. Includes bibliographical references (leaves 155-162)

M.Com. (Informatics) / A certain degree of vulnerability exists in traditional knowledge-based identification and authentication access control, as a result of password interception and social engineering techniques. This vulnerability has warranted the exploration of additional identification and authentication approaches such as physical token-based systems and biometrics. Speaker recognition is one such biometric approach that is currently not widely used due to its inherent technological challenges, as well as a scarcity of comprehensive literature and complete open-source projects. This makes it challenging for anyone who wishes to study, develop and improve upon speaker recognition for identification and authentication. In this dissertation, we condense some of the available speaker recognition literature in a manner that would provide a comprehensive overall picture of speaker identification and authentication to a wider range of interested audiences. A speaker recognition solution in the form of an open, user-friendly software prototype environment is presented, called SRIA (Speaker Recognition Identification Authentication). In SRIA, real users may enrol and perform speaker identification and authentication tasks. SRIA is intended as platform for speaker recognition understanding and further research and development.

No description available.

published_or_final_version / Computer Science / Master / Master of Philosophy

HMM-based systems for Automatic Speech Recognition typically model the acoustic features using mixtures of multivariate Gaussians. In this thesis, we consider the problem of learning a suitable covariance matrix for each Gaussian. A variety of schemes have been proposed for controlling the number of covariance parameters per Gaussian, and studies have shown that in general, the greater the number of parameters used in the models, the better the recognition performance. We therefore investigate systems with full covariance Gaussians. However, in this case, the obvious choice of parameters – given by the sample covariance matrix – leads to matrices that are poorly-conditioned, and do not generalise well to unseen test data. The problem is particularly acute when the amount of training data is limited. We propose two solutions to this problem: firstly, we impose the requirement that each matrix should take the form of a Gaussian graphical model, and introduce a method for learning the parameters and the model structure simultaneously. Secondly, we explain how an alternative estimator, the shrinkage estimator, is preferable to the standard maximum likelihood estimator, and derive formulae for the optimal shrinkage intensity within the context of a Gaussian mixture model. We show how this relates to the use of a diagonal covariance smoothing prior. We compare the effectiveness of these techniques to standard methods on a phone recognition task where the quantity of training data is artificially constrained. We then investigate the performance of the shrinkage estimator on a large-vocabulary conversational telephone speech recognition task. Discriminative training techniques can be used to compensate for the invalidity of the model correctness assumption underpinning maximum likelihood estimation. On the large-vocabulary task, we use discriminative training of the full covariance models and diagonal priors to yield improved recognition performance.

This thesis presents a cloud platform for automatic speech recognition, CloudASR, built on top of Kaldi speech recognition toolkit. The platform sup- ports both batch and online speech recognition mode and it has an annotation interface for transcription of the submitted recordings. The key features of the platform are scalability, customizability and easy deployment. Benchmarks of the platform show that the platform achieves comparable performance with Google Speech API in terms of latency and it can achieve better accuracy on limited domains. Furthermore, the benchmarks show that the platform is able to handle more than 1000 parallel requests given enough computational resources. 1