Artificial Neural Network-based Speech Recognition Studies

Language is one of the most important means of exchanging information among themankind. Since the computer was invented, many scientists have been devoted to enablingthe computer to understand the natural language. Speech recognition is a comprehensivetechnology of such areas as acoustics, phonetics, linguistics, computer science, informationprocessing and artificial intelligence, which can be used widely. The research of speechrecognition technology has been focused by the world for a long time.The neural network is a new developing science, which simulates the mechanism ofhuman brain and was putted forward by the developing of modern science. It is not theoverall description of human brain, but the abstract, simulation and simplifying of thephysical neural networks of human beings. The purpose of the research in this area isexploring the human brain mechanisms in information processing, storing and searching. Ifpeople can understand these mechanisms, a new way for the research of artificialintelligence, information processing and etc can be opened up. Artificial neural network isa system which using a physically feasible system to imitate the structure and function ofnerve cells in human brain, which has the ability of self-learning, contrasting, reasoningand summarizing It have offered a new way in solving such complicated patternclassification problems as speech recognition. The characteristics of speech and the setupof neural network model were researched deeply in this dissertation.1. The developments in the areas of speech recognition and artificial neural network were summarized and analyzed. The important technologies in speech recognition such as the classification of speech recognition technologies, the composition of the speech recognition system, pre-processing of speech signals and the feature extraction of speech signals (LPCC, MFCC) were researched.