连呼数字的分割是连呼数字识别需解决的主要问题之一。本文在用短时能量、短时过零率等参数对汉语数字语音统计分析的基础上,针对其参数特点,提出了一种多层次的汉语连呼数字分割算法,并在IBM-PC机上进行了模拟分割,取得了较好的分割结果。 One of the main problems that need to be solved is to divide the number of consecutive calls into two. Based on the statistical analysis of Chinese digital speech with parameters such as short-term energy and short-term zero-crossing rate, aiming at the characteristics of its parameters, a multilevel Chinese L2 digital segmentation algorithm was proposed and implemented on IBM-PC A simulation of segmentation, and achieved a better segmentation results.