Secondly , according to the model of speech - recognition system it describes the fundamental of every part ( including pretreatment , end point detection , template production and template matching ) such as the speech data signal process , the mode recognition and discusses some method of the realization of speech - recognition on pc 其次,根据语音识别系统的基本构成模型,介绍了预处理、端点检测到模板生成及模板管理和模板匹配各部分所涉及到的语音数字信号处理、模式识别等方面的基本原理,并讨论了在pc机上实现语音识别系统的几种途径。