The Speech Recognition System Research And Design Base On Isolated-word In Small-vocabulary

Using speech signal as research object, speech recognition is an important research direction of speech signal processing and it is an embranchment of pattern recognition. With the progress of modern computer technology, the people use computer more and more. Communicating through speech with computer is the most comfort way, so the speech recognition become an important research field of modern computer technology.In this paper, the speech recognition system for a basic introduction, and discussed a based on the small isolated word speech recognition system to achieve. First we analyse some basic concept and theory about speech recognition system. Then according to the algorithms of the end-point detection for pretreatment, feature extraction of speech recognition, pattern matching and model training, we have a deep study on them. Using the "double threshold" law we can complete the end-point detection of the pronunciation, and obtain the more ideal examination effect under the laboratory environment. Discuss the methods of extracting the different main characteristic parameters of speech recognition systematically, especially analyze LPC and MFCC parameters of the characteristic parameters, Aiming the pattern matching and the model training question, this article propose the speech recognition model based on the CDHMM. The experiment proved that under the signal to noise ratio lower laboratory environment, this model has obtained the good recognition result.