Design And Implementation Of Voice Control System Based On Android Platform

With the development of mobile communication technology and popularization of computer application, communication based on natural language between users and mobile terminal has become one of the most convenient interactive approach. The combination of the speech recognition technology and the real-time communication of the intelligent control technology technique is considered as the MIVAS with the widest market prospects and development potential, because it can not only meet the need for a more efficient information transmission in modern society, but also make entertainment more available.This thesis examine the research status and development prospects of the voice control system, analyze the current market demands, design and accomplish a voice control system based on Android platform. With the study of open source toolkit Lucence and the key technology in Chinese word segmentation, this thesis propose a combination between string matching technology and statistical points lexical for Chinese word segmentation technology. The technology will forward maximum matching method combined with high-frequency lexical priority points, the first to use high frequency priority to short term lexical word processing, and then combined with the context of positive maximum matching, this technology has overcome the disadvantages of traditional positive maximum matching method which matching word by word. This thesis used finite state machine to control the background status conversion. Designed a finite state machine is used to control the state transitions between various modules, and gives the most simplified the proof of the state machine.This thesis develops the voice control system which utilizes IFLYTEK ASR (Auto researched and Recognize, automatic Speech recognition) interface for front-end Speech recognition and synthesize and applies Lucence kits for Chinese word segmentation processing application for the background. Also, it uses the finite state machine to control system background state transition. And enables users can control all various of the applications on mobile terminal quickly and conveniently through simple voice command or order words, especially for some environment as in car, in the bathroom and in hospital environment to control mobile terminals, etc. In addition, this thesis takes a real-time interactive application software which is similar to Wechat as an example to finish the system-test which suggests that the system has a strong ability of speech recognition and intelligent control, also, it has sensitive reaction which satisfies the need of users for effective, efficient and convenient man-machine interaction.