Research Of The Speech Recognition Technology Based On HMM

Speech recognition is very promising in application. As an interdisciplinary field, it is also theoretically very valued. In this paper, considering existent technology of speaker recognition, the author study the speaker recognition method based on SOFMNN and HMM, and set up speaker's recognition system.This dissertation has analyzed since pretreatment of the signal of the pronunciation, measure the extreme point to the signal of the pronunciation, filter except silent section and noise of the signal of the pronunciation, has offered the effective pronunciation section for abstraction of the characteristic parameter of the pronunciation. Traditional pronunciation extreme point detection method and on the basis of LPCMFCC extreme point performance of detection method have relatively also in the article, draw the conclusion: On the basis of LPCMFCC extreme point detection method can finely appear pronunciation come by extreme point to measure under the environment of high noising.Theoretical model is significant to help design the speech recognition system. This paper has analyzed the speech recognition based on HMM, realize the process of the isolate and continuous recognition. Then, according to the characteristic of SOFMNN, this paper tries to use the mixed model in the speech recognition.At the end of this paper, presents speech recognition achieved on MATLAB platform, including the principle of realization, the selection of speech signal, endpoint detection. By comparing the mixed and the single model, we know that this mixed model has better anti-noise robustness.