基于hmm的汉语数字语音识别系统研究-电路与系统专业论文.docxVIP

ABS ABSTRACT Speech i s an important tool for people communicating with machines．The technology of Automatic Speech Recognition(ASR) p can let machines understand humanity’S language and carry out corresponding operations and it can be widely used in many areas． Though it has been studi ed for many years，there are st i i i many problems in ASR worthy to be studied． The Acoust i c mode l of the speech——product i on／ speech——percept and the speech recognition theory are the basics of ASR，SO each step of ASR process i S analyzed in detai Is．The improved spectrum entropy algorithm i s brought forward for the endpoints detection，and the results of experiments show that robustness of the system has been improved while using this method for endpoints detect ion．The chosen speech feature parameters have great effects on robustness and real——time of the speech recognition system．After introducing short—time feature parameters and spectrograms，three approaches to extracting speech feature parameters such as Linear Predictive Coding(LPC)，Linear Predictive Cepstrum Coefficients(LPCC) and Mel—Frequency Cepstrum Coefficients(MFCC)，are discussed in detai 1 S，then their di stort ion measure are introduced． The DTW theory and HMM theory are di scussed．Their applications in recognition are analyzed through the MATLAB ．1 programs．The i solated word recogni t ion systems based on the DTW theory for speaker。—independent and speaker dependent are discussed whi le using different feature parameters．0n the other side，the small—vocabulary speaker—independent speech recognition recognition system based on HMM iS constructed．Different feature parameters can be chosen in the recongnition system， which has good robustness．The experiments are conducted to recognize the mandarin digitals from 0 to 9 with this system． The results show that 12——dimension LPCC is the most effective feature，but the recognition rate of 26一dimension MFCC is about 10％higher than that of 12一dime