基于神经网络的语音合成环节中英语单词音节划分的技术  

Neural Network Based Syllable Boundary Assignment for Letter-to-sound of English Word

作　　者：张浩然[1] 韩正之[1] 李昌刚[1] 

机构地区：[1]上海交通大学自动化系,上海200030

出　　处：《计算机工程与应用》2003年第9期95-97,共3页Computer Engineering and Applications

摘　　要：提出了用神经网络来划分英语单词内音节的新方法,该方法比传统的查表法具有很大的优点,如计算速度快,内存需求量少,具有一定的泛化能力,实验证明文章提出方法的正确性和有效性。This paper develops a new approach which is based on neural network technique to separate syllable boundary of English word.The approach is(better than)superior to traditional dictionary look-up method such as its low computation complexity,requiring less memory and possessing generality ability.The experiment demonstrates the correctness and effectiveness of proposed strategy.

关 键 词：语音合成 音节划分 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]