【摘要】 在语音识别系统中,Trigger模型作为语言模型的一种,用于描述长距离词与词之间的关系,然而以往的Trigger语言模型多是针对单个词的模型,本文借鉴数据挖掘中关联规则发现的Apriori算法,利用效率比较高的FP树算法产生多词Trigger对,由此构造多词Trigger对语言模型,这种模型能够更多地利用历史数据,弥补了传统N元文法语言模型描述距离小于N的缺点.更多还原

【Abstract】 Trigger pair language model has been used to investigate long distance dependent relationship for speech recognition systems.Previous trigger pair model has only one word for its trigger.In this paper,a multiple words trigger pair model was created by using Apriori algorithm of mining association rules in a large database and more efficient FP-tree algorithm.This new model can know more about the history for a better prediction of the current word and can overcome the shortcoming of N-gram model that it only describe the word pairs being less than N words apart.更多还原