No description available.

This thesis presents two novel methods for isolated word speech recognition based on sub-word components. A digital neural network is the fundamental processing strategy in both methods. The first design is based on the 'Separate Segmentation & Labelling' (SS&L) approach. The spectral data of the input utterance is first segmented into phoneme-like units which are then time normalised by linear time normalisation. The neural network labels the time-normalised phoneme-like segments 78.36% recognition accuracy is achieved for the phoneme-like unit. In the second design, no time normalisation is required. After segmentation, recognition is performed by classifying the data in a window as it is slid one frame at a time, from the start to the end of of each phoneme-like segment in the utterance. 73.97% recognition accuracy for the phoneme-like unit is achieved in this application. The parameters of the neural net have been optimised for maximum recognition performance. A segmentation strategy using the sum of the difference in filterbank channel energy over successive spectra produced 80.27% correct segmentation of isolated utterances into phoneme-like units. A linguistic processor based on that of Kashyap & Mittal [84] enables 93.11% and 93.49% word recognition accuracy to be achieved for the SS&L and 'Sliding Window' recognisers respectively. The linguistic processor has been redesigned to make it portable so that it can be easily applied to any phoneme based isolated word speech recogniser.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Many complex diseases are known to have a substantial genetically heritable component. Elucidation of these genetic risk factors provides increased knowledge of the biological mechanisms that result in the diseases while also presenting new potential targets for therapy. This thesis explores the methodology of mapping genetic loci using isolated populations in the context of quantitative trait analysis. Chapter 1 explores the rational for the project, discussing the benefits of using quantitative traits rather than binary disease status and the pros and cons of using isolated populations. This is followed by a brief history of genetic mapping with reference to type 2 diabetes mellitus (T2D) and related quantitative traits. Chapter 2 introduces the methods used in this thesis. This includes strategies to deal with medication, methods to determine kinship between individuals, linkage analysis, association analysis and meta‐analysis of multiple studies. Chapter 3 presents linkage analysis of T2D related traits carried out in 2 – 4 populations depending on availability of the traits and appropriate marker data. Chapter 4 presents the results of association analysis for T2D related traits in 3 – 5 populations using genome‐wide SNP data. The results using the alternate methods described in chapter 2 are compared using fasting glucose as this was the most widely measured phenotype. Chapter 5 introduces additional traits derived by pulse wave analysis and discusses their relevance to metabolic disease before presenting association analysis using the preferred method from chapter 4. An overall discussion of the strengths and weaknesses of the analysis is given in chapter 6.

The purposes of this study were: (1) to develop a program based upon a combination of previously tested techniques, (2) to adapt these techniques for use by school personnel within the classroom situation, (3) to test this program upon an all-black, multi-age kindergarten and a first-grade classroom in an inner-city school, and (4) to evaluate the effectiveness of this program.