方言语音识别技术,即汉语方言地理信息系统平台,就是根据人的声音地域特征开展身份比对的方式。同指纹一样,语音的地域特征是难以改变的,属于人的个性生理特征之一,在现实生活中很难找到一个人能改变十岁以前形成的方言特征(即母语腔)。随着方言语音比对技术的日趋成熟和完善,其识别技术以独特的优势日益显现。本文通过介绍方言语音识别的概念、原理及其识别方法,分析方言语音识别技术在公安工作中的应用及其前景。 Dialect speech recognition technology, that is, the Chinese dialect geographical information system platform, is based on the geographical characteristics of people’s voice to carry out identity comparison. As with fingerprints, the geographical features of speech are hard to change and belong to one of the individual physiological characteristics of a person. It is very hard to find in one’s life how one can change the dialect characteristics (ie mother tongue cavity) formed before the age of ten. With the dialect voice comparison technology matures and perfected, its identification technology is increasingly apparent with unique advantages. This paper introduces the concept and principle of dialect speech recognition and its identification method, and analyzes the application of dialect speech recognition technology in public security work and its prospects.