维吾尔语语言合成系统前端文本处理技术研究

摘要

文本分析、特殊符号处理、韵律短语划分是语音合成系统前端处理模块的核心内容.根据维吾尔语的语言和语音特征,提出一套准确可行的前端文本处理策略,很好地分析与描述了维语语言层的信息,并将之映射到语音层上.研究的完成为进一步实现维吾尔语语音合成系统打下坚实的基础.%Text analysis, special symbols processing and rhythm phrases division are the core content of the front part processing module of the speech synthesis systems. This paper proposes a set of accurate and feasible front part text processing strategy according to Uighur language features, makes a good analyses and description of the information of Uighur language layer and mapped the information layer to voice layer. The completion of this research establishes a solid foundation for the Uighur speech synthesis system further.