Search

This paper examined the speech sound errors in the sentences of five adult Japanese English Langue Learners (ELLs) and classified the errors as phonological processes. The processes observed were compared to the established developmental processes exhibited by monolingual English speaking children to provide insight into the similarities between L1 phonological acquisition and L2 phonological learning. The results revealed that a majority of the incorrect phonemes are also phonemes that are mastered late in the development of English speaking children. A great deal of variation in rate of occurrence of different phonological processes existed across speakers. Rounding, tensing, decentralization, vowelization, final devoicing, and cluster reduction were the most frequently occurring vowel and consonant processes. The results of this study are interpreted in relation to markedness and a contrastive analysis between Japanese and English phonetic inventories. Clinical implications with respect to pronunciation instruction for ELLs are also addressed.

English is a vehicle for international communication. In order to meet the demands of modern society, English teachers need to pay more attention to the development of learners’ competence and focus on a more effective and successful method. However traditional approaches to English language teaching still dominate Thai classrooms. Language teachers should not focus on reciting but should teach from their own understanding of language learning and help learners gain more competence with confidence. This study is a collaborative action research investigation to develop pronunciation training and communicative competence for Thai students studying English in Thailand. This study investigated pronunciation training and language learning strategies, how they influenced the learning behaviour of Thai students studying English and improved their speaking confidence. The purpose of the training was to improve students’ pronunciation and spoken intelligibility. It drew upon data collected in pronunciation training in one school in Thailand using language learning strategies and evaluated improvement after being trained in developing speaking confidence. The project contained two cycles, the first of which was to train five teachers using pronunciation training and language learning strategies. We evaluated their improvement in correct speech and in developing speaking confidence. In the second cycle, these teachers in turn taught a group of four students each and similar improvements were observed. The action phases showed the implications of the importance of pronunciation training in the Thai context and the usefulness of dictionary usage to help learners to improve their competence and to have more confidence to speak English. The project resulted in a change of policy by the school to include pronunciation teaching and to allocate English classes to teachers who understood that process. The pronunciation learning strategies in this study and those of other researchers were presented to formulate strategies as a contribution for teachers to include teaching pronunciation in their classroom instruction. The researcher intends that the data will be useful for language teachers to help them further their understanding of their students’ learning behaviour to achieve improved pronunciation. In addition, the phonetic symbol system used in the training was chiefly inspired from the symbols of the International Phonetic Association (IPA) to be standardized and easy to apply.

This thesis describes the development of unsupervised neural-based pattern classifiers for the training of vowel pronunciations for students learning a foreign language. This paper examines American learners of the French language. A corpus of single word utterances is compiled from a group of native French speakers. Cepstral features are used to train two unsupervised neural pattern classifiers: a self-organizing map and a growing neural gas. The development and justification for the use of these classifiers is presented. The output from the classifier is translated to a bar graph for visual assessment. The degree to which the utterance sounds native is determined by comparing target graphs and those produced by the user. It is concluded that unsupervised classification techniques can be used to develop a pronunciation training system that is independent of the language used to train the system. This allows for pronunciation training to be easily achieved for low-density languages.