TTS在办公系统的研究与开发

发布时间：2018-05-20 11:18

本文选题：合成 + 识别　；参考：《青海师范大学》2012年硕士论文

【摘要】：语音技术是将近半个世纪发展起来的一门新型科学技术，包括语音识别和语音合成。语音识别的目的是让机器把语音信号转化成可以接受的文本文件或者控制命令。语音合成则是将文本转化为人类可以理解的声音信号。语音技术在工业、军事、交通、医学、民用等各方面有广阔的应用前景，蕴含有巨大的社会经济效益，受到各国日益重视。其中，文字-语音的转换（简称TTS）系统是一项非常有用的语音技术，它可以将任意文本信息转换为语音信号。它的用途很广泛，如汽车导航，电信的呼叫服务，公交车到站站名自动播报等。现有的基于大语料库的TTS系统已经能够产生高清晰度和自然度的合成语音，并已得到较大的发展和应用。文本转换成语音仍是当前国内外语音方面的一个热点，其核心主要是语音库的建立与搜索引擎的实现与优化问题。因语音在生活中时刻都存在着，我们经常用语音直接打交道，就是根据语音与文本之间存在一定的规律去发现与开发新的产品来为人民之利益，改善大众生活。目前对于语音在办公软件方面的应用还很欠缺。本文主要研究TTS在办公应用软件中的开发与应用，以使办公“声音”化，，为社会中的一些渴求知识的残疾人找到了“新出路”。本文主要阐述通过利用微软提供的Speech SDK及Agent技术，在办公应用软件中实现语音的识别和合成。利用Speech SDK的API接口和Agent的语音识别方法，可以将办公应用软件中的文本依次朗读出来。同时也可以通过语音识别来进行语音留言，更加方便用户的使用。利用其TTS技术，可以实现语音的合成，能进行中英文混合阅读。通过设置Agent的属性设置与应用，还可以进行动态效果显示。本论文的研究对于TTS技术应用在社会中的经济实用价值取向有很大的推动作用。
[Abstract]:Speech technology is a new science and technology developed in nearly half a century, including speech recognition and speech synthesis. The purpose of speech recognition is to allow machines to convert speech signals into acceptable text files or control commands. Speech synthesis is the translation of text into sound signals that humans can understand. Speech technology has a broad application prospect in industry, military, transportation, medicine, civilian and so on. It contains enormous social and economic benefits, and has been paid more and more attention by many countries. Among them, text-speech conversion (TTS) system is a very useful speech technology, it can transform any text information into speech signal. It has a wide range of applications, such as car navigation, telecommunications call service, bus stop name automatic broadcast, etc. The existing TTS system based on large corpus has been able to produce high definition and naturalness speech synthesis, and has been developed and applied. Text conversion to speech is still a hot topic at home and abroad, the core of which is the establishment of phonetic corpus and the realization and optimization of search engine. As speech exists all the time in our lives, we often deal with it directly, that is, to discover and develop new products for the benefit of the people and to improve the life of the masses according to the certain rules between the speech and the text. At present, the application of voice in office software is still very deficient. This paper mainly studies the development and application of TTS in office application software, in order to make the office "sound" and find a "new way" for some disabled people who want knowledge in the society. This paper mainly describes how to realize speech recognition and synthesis in office application software by using Speech SDK and Agent technology provided by Microsoft. By using the API interface of Speech SDK and the speech recognition method of Agent, the text in office application software can be read out in turn. At the same time, voice recognition can also be used to voice message, more convenient for users to use. By using TTS technology, speech synthesis can be realized and mixed reading between Chinese and English can be carried out. By setting the properties of Agent and the application, we can also display the dynamic effect. The research in this paper has a great impact on the economic and practical value orientation of the application of TTS technology in society.
【学位授予单位】：青海师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TN912.3

【参考文献】