摘要
文章讨论了一个典型的汉语文语转换系统的实现。首先介绍了系统的整体框架及其各个功能模块，然后分析了系统的特点及其存在的问题，最后从两方面讨论了改进系统的具体思路：提出了一种简单而有效的基音周期提取算法，验证了上下文环境在提高合成语音自然度中的作用。
This paper discusses the implementation of a typical Chinese Text-to-Speech op) system. Firstly,It presents the frame structure of the system and discusses all of the modules in detail,then analyzes its characteristics and problems existing. Lastly This paper does some experiments in order to improve the quality of the system: It puts forward a simple and effective algorithm to calculate the base period Of speech waveforms and verifis the importance of context in improving the naturalness of synthesis speech.
出处
《计算机工程与应用》
CSCD
北大核心
2000年第9期1-3,共3页
Computer Engineering and Applications
基金
国家自然科学基金!(编号为69523001)的部分资助
