A Model of Text-to-Speech Application for Learning Thai Language

摘要

Language is the ability to express one's thoughts by means of a set of signs, whether graphical, acoustical or even musical. It is a distinctive feature of human beings, who are the only creatures to use a structured system of language. Speech is one of its main components. It is by far the oldest means of communication between people and it is also the most widely used. Conversation dialogs are one of the important foundations for the practical use of any language. In the past, a learner could learn the correct pronunciation from tapes that come with the books or training materials. However, the learners could learn to pronounce only those words in limited vocabulary of the training material. When learners find new words that do not exist in the book or training materials, they must guess how to pronounce them. In the case of the Thai language, the pronunciation of each word becomes more difficult because the Thai language is a tonal language. This means that the pitches or tones of spoken words are meaningful. A word pronounced with different pitches carry different meanings. Therefore, Text-to-Speech synthesis can be coupled with computer aided learning systems to provide a helpful tool to learn a language. When designing a voice based application, it is important to understand the limitations of the text-to-speech component and ensure you design around or generate the grammars needed to help ensure that these limitations have a minimal impact on the successful use of the application. This paper presents guidelines and an example of how to design a text-to-speech application for helping foreigners learn the Thai language.