基于嵌入式系统的特定说话人语音识别系统的实现

The Implementation of Special Speaker's Speech Recognition System Based on Embedded System

语音识别是实现人机通信的一种重要技术手段.本文基于Samsung公司开发的B lues芯片,应用隐马尔可夫模型(HMM)理论和方法,完成了针对特定人、小词汇量嵌入式语音识别系统开发.文章介绍了语音识别过程的各个环节及其实现方法.

Speech recognition is an important technical means of man-machine communication system. An embedded recognition system based on the hidden Markov model （ HMM ） theory is developed using the chip Blues produced by Samsung. The result of experiments shows that the model applied to this system performed very well, could successfully recognize small amount of words of certain speakers with considerable high recognition efficiency. In this thesis, firstly, a brief introduction of the chip Bluse is given. Secondly, the speech recognition processes and its realization method, （ especially of the basic theory of hidden Markov model） are introduced.