Research And Implementation Of Chinese Speech Recognition Based On Speaker-dependent

In recent years, with the furthering research on speech recognition theory and the development of software and hardware technology for digital signal processing, the research of speech recognition technology is becoming a study focus. In addition, with the development of embedded technology, embedded speech recognition technology has become an important development direction in speech recognition field.This paper makes a briefly introduction to the basic knowledge of speech recognition technology, presents a system model of speech recognition and mainly conducts a study on algorisms involved in the model. Furthermore, this paper also researches and implements an embedded speech recognition system, and gives a prospect about the development of speech recognition technology.The focus of the paper is the research and implementations of speech recognition algorisms. According to the basic model of the speech recognition system, this paper does an intensive study on speech signal preprocessing, feature extraction and recognition algorithms. This paper also implements a person-dependent, small vocabulary, isolated word speech recognition system on PCs with MATLAB language and the dynamic time warping algorithm and analyzes the testing results. Furthermore, this paper implements the speech recognition system in C language and applys it to XSBase development platform and makes a research on embedded speech recognition technology.