汉语语音合成——原理和技术

• 作 者: 籍顺心
• 机 构: 声学学报,
• 出 处: 声学学报
• 发表日期: 2012年03期
• 关键词: 汉语语音合成 语言声学 会说话 声学研究 马先生 大猷 语音合成技术 声学特征 国语 语音识别

创造一个会说话的机器不仅是人类早就有的梦想,也是人类孜孜以求的目标之一。而“语音合成”就是一种用机械的或电子的手段,产生人造语音的技术。多年来各国科学家为此作了不懈的努力,并取得了丰硕的成果。语言声学是我国声学研究长期坚持的重要领域之一。早在20世纪40年代,马大猷教授就有“国语中语音分配”的论文发表。新中国成立后,马先生在筹备组建中国科学院声学研究所的过程中,就专门设立了

Creating a talking machine is not only an early human dream, but also one of the goals pursued by mankind. And “voice synthesis ” is a kind of mechanical or electronic means to produce artificial voice technology. Over the years, scientists from all countries have made unremitting efforts to this end and have achieved fruitful results. Speech acoustics is one of the important fields that our acoustics research insists on for a long time. As early as the 1940s, Professor Ma Danyu published an article entitled “Voice Distribution in Mandarin”. After the founding of New China, Mr. Ma specially set up the process of preparing the Institute of Acoustics of Chinese Academy of Sciences