一种多层次的汉语连呼数字语音分割算法

A New Multi-Layer Segmentation Algorithm for Connected Digital Speech

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。

There is an important problem which must be solved in connected digit recognition, i.e. the segmentation. In this paper, a multi-layer segmentation algorithm is proposed which is based on the statistical feature analysis of the digital speech in Chinese using energy and zero-cross parameters. For strings of variable length from 3—4 digits, preliminary results show that segmentation algorithm is effective and feasible in the given conditions.