%0 Journal Article
%T 基于得分归一化和系统融合的语音关键词检测方法<br>Keyword Spotting based on Score Normalization and System Combination
%A 李鹏
%A 屈丹
%J 数据采集与处理
%D 2017
%R 10.16337/j.1004-9037.2017.02.016
%X 为了有效利用不同关键词检测系统的互补性，解决不同系统检测结果置信度得分不在同一范围的问题，提出了一种基于得分规整和系统融合的语音关键词检测方法。首先，为了克服连续语音识别系统中因剪枝错误而引起的关键词丢失问题，应用了关键词相关的软Beam宽度剪枝策略裁剪词图；其次，在系统融合前采用得分归一化方法，使得不同系统关键词检测结果置信度得分在同一范围；最后，通过系统融合处理将不同系统的关键词输出进行整合，得到最终的关键词检测结果。实验结果表明，经过得分归一化处理后，关键词检测性能的实际查询词权重代价(Actual term-weighted value, ATWV)平均相对提升30%；系统融合后关键词的检测性能，相比于得分归一化处理后的最佳单一系统，得到了10%的提升。<br>To effectively use the complementarity of different keyword spotting systems and solve the problem that the confidence scores from several different subsystems is not in the same range, a keyword spotting system based on score normalization and system combination is proposed. Firstly, to avoid keyword missing due to pruning errors in a large vocabulary recognition system, the keyword soft Beam pruning method is presented. Secondly, score normalization is needed to transform these confidence scores into a common domain, prior to combining them. Finally, after score normalization，the outputs are combined from different subsystems. Results show that score normalization methodology improves keyword search performance by 30% in average. Experiment also show that combining the outputs of diverse systems, system perform is 10% better than the best normalized KWS system.
%K 关键词检测
%K 得分归一化
%K 系统融合
%K 软Beam剪枝&lt
%K br&gt
%K keyword spotting
%K score normalization
%K system combination
%K soft Beam pruning
%U http://sjcj.nuaa.edu.cn/ch/reader/view_abstract.aspx?file_no=201702016&flag=1