Research On Embedded Speech Synthesis Technology

Speech synthesis is an information processing technology that turns the text information into standard smooth voice information. Speech synthesis technology has been proposed for hundred years. After hundred years of research and development, speech synthesis technology has made tremendous progress on the synthesis of philosophy and the synthesis algorithm which can be realized. The technology has wide application in the field of human-computer interaction and text information processing.Firstly this paper introduces the development of speech synthesis technology, and what level it is. Describing the various methods of speech synthesis technology, for example:direct simulation vocalization formant speech synthesis, LPC synthesis, PSOLA algorithm. Introduce the knowledge of Chinese speech and text content standardized process which play an important role in speech synthesis technology. In order to establish the text to speech conversion system which is suitable for running on embedded systems, this paper describes that how to establish a speech database, as well as the process of how to do with polyphone, and the various components and production process of the voice index module and voice playback module.In addition, this paper produces a text to speech conversion system which has text to speech conversion function based on own voice library, the voice index dynamic link library, Play voice dynamic link library, which are achieved by C++programming language and do not call the third-party components.