语音查询项检索中的两阶段得分规整方法  

Two-Stage Score Normalization Method for Spoken Term Detection

作　　者：李鹏[1] 屈丹[1] 

机构地区：[1]解放军信息工程大学信息系统工程学院,郑州450001

出　　处：《模式识别与人工智能》2016年第3期216-222,共7页Pattern Recognition and Artificial Intelligence

基　　金：国家自然科学基金项目(No.61403415;61175017)资助~~

摘　　要：得分规整为语音查询项检索系统中的必要过程,文中提出两阶段得分规整方法.先引入rank-p和relativeto-max这2个特征至区分性得分规整方法中,使正确候选结果和错误候选结果的置信度得分区分性更大,更易进行关键词确认.再应用基于优化查询项权重代价指标的得分规整方法得到最优的语音查询项检索性能.实验表明,文中方法同时利用区分性和基于优化查询项权重代价指标得分规整方法的优点,相比最佳单一得分规整方法性能更优.Score normalization is an essential part for a spoken term detection （STD） system. In this paper, a two-stage score normalization method is proposed. Firstly, two features, rank-p and relative-to-max are introduced into a discriminative score normalization method to get more discriminative confidence scores between correct and wrong candidate words, and thus the keyword verification is more accurate. Secondly, a term-weighted value evaluation metric based normalization method is applied to further optimize the performance of STD. Experimental results show that the proposed method takes advantages of both discrimination and metric-based score normalization methods, and it obtains better performance than the best single score normalization method does.

关 键 词：语音查询项检索 得分规整 区分性模型 置信度得分 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]