Development And Research Of Chinese Isolated-speech-recognition System

The Speech Recognition System has become more and more perfect through its development for almost half of a century. But, Chinese Speech Recognition is not ideal and there is a long distance for its application owing to the problems such as complicate Chinese structure, a large number of single syllable, same syllable, dialect and anomalous grammar. The paper analyses and explains particularly about basic principles, applied methods and a part of actual technical details.Firstly, this paper introduces definition and meaning, situation of development both at home and abroad; specifies the basic principles of Speech Recognition simply.Secondly, according to the constitutive model of basic principles of Speech Recognition System, the paper introduces the basic principles about the speech data signal process and the mode recognition and so forth from pretreatment, end point detection to template production, template management, template match; debates some methods of realization of Speech Recognition System on PC.Then, this paper introduces a simple Speech Recognition System of speaker independent, small vocabulary, and isolated-speech-recognition system by using the Visual C++6.0 according to the DTW model's Speech Recognition exercises and basic methods. The basic constitutive model between system and Speech Recognition is same. In the exercising process, we can read the sample data from the wave file; calculate the character vector made by 12 ranks of LPC and 12 ranks of LPCC; exercise according to the assembled methods; gain the necessary templates of sequential Speech Recognition; lay them to the template library. While recognizing, it is important to calculate the character vector of speech input, gain the best template match or make refuse judgment by contrasting the template match's methods and the speech template in the template library.Finally, this paper summarizes the system's function and indicates its improvable direction. It is a basic research for us to further develop the practicality of Speech Recognition System through actual test and study of Speech Recognition System.