基于加权有限状态机的动态匹配词图生成算法

doi:10.3724/SP.J.1146.2013.00422

摘要
图/表
参考文献
相关文章 (1)

全文: PDF (292 KB)
输出: BibTeX | EndNote (RIS)

摘要由于现有的加权有限状态机(WFST)解码网络没有精确词尾标记，导致当前已有的词图生成算法不含精确的词尾时间点，或者仅是状态、音素级别的词图，无法应用到关键词检索中。该文提出在WFST静态解码器下的语音识别词图生成算法。首先从理论上分析了WFST解码音素图和词图的可转换关系，然后提出了字典的动态音素匹配方法解决了WFST网络中词尾时间点对齐的问题，最后通过令牌传递的遍历方法生成了词图。同时，考虑到计算量优化，在令牌传递过程中引入了剪枝算法，使音素图转词图的耗时不到解码耗时的3%。得到的词图，不仅可以用于语言模型重打分，由于含有精确的词尾时间点，还可以直接应用到关键词检索系统中。实验结果表明，该文的词图生成算法具有较高的计算效率；和已有动态解码器的词图相比，词图中包含更多解码信息，在大词汇连续语音识别的重打分结果和关键词检索中都能取得更好的性能。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	郭宇弘
	黎塔
	肖业鸣
	潘接林
	颜永红

关键词 ：自动语音识别, 加权有限状态机, 词图生成, 关键词检索

Abstract：The existing lattice generation algorithms have no exact word end time because the Weighted Finite State Transducer (WFST) decoding networks have no word end node. An algorithm is proposed to generate the standard speech recognition lattice within the WFST decoding framework. The lattices which have no exact word end time can not be used in the keyword spotting system. In this paper, the transformation relationship between WFST phone lattices and standard word lattice is firstly studied. Afterward, a dynamic lexicon matching method is proposed to get back the word end time. Finally, a token passing method is proposed to transform the phone lattices into standard word lattices. A prune strategy is also proposed to accelerate the token passing process, which decreases the transforming time to less than 3% additional computation time above one-pass decoding. The lattices generated by the proposed algorithm can be used in not only the language model rescoring but also the keyword spotting systems. The experimental results show that the proposed algorithm is efficient for practical application and the lattices generated by the proposed algorithm have more information than the lattices generated by the comparative dynamic decoder. This algorithm has a good performance in language model rescoring and keyword spotting.

Key words： Automatic speech recognition Weighted Finite State Transducer (WFST) Lattice generation Keyword spotting

收稿日期: 2013-04-01

PACS:

TP391.42

基金资助:

国家自然科学基金(10925419, 90920302, 61072124, 11074275, 11161140319, 91120001, 61271426)，中国科学院战略性先导科技专项(XDA06030100, XDA06030500)，国家863计划项目(2012AA012503)和中科院重点部署项目(KGZD-EW-103-2)资助课题

通讯作者: 郭宇弘 E-mail: guoyuhong@hccl.ioa.ac.cn

引用本文:

郭宇弘, 黎塔, 肖业鸣, 潘接林, 颜永红. 基于加权有限状态机的动态匹配词图生成算法[J]. 电子与信息学报, 2014, 36(1): 140-146. Guo Yu-Hong, Li Ta, Xiao Ye-Ming, Pan Jie-Lin, Yan Yong-Hong. Exact Word Lattice Generation in Weighted Finite State Transducer Framework. , 2014, 36(1): 140-146.

链接本文:

http://jeit.ie.ac.cn/CN/10.3724/SP.J.1146.2013.00422 或 http://jeit.ie.ac.cn/CN/Y2014/V36/I1/140