本文对计算机语音命令理解的算法作了一些探索性的研究。首先针对词图结构的特点提出了一种词图树扩展理解算法,通过分析与实验比较,发现该算法在保证精确率的下降很小的条件下可获得比传统的Nbest路径理解算法高得多的召回率,而计算效率仅相当于Nbest路径理解算法中句子候选数取值很小时的情况;其次根据对实验结果的分析与观察,给出了一种行之有效的命令理解容错算法,使得理解召回率提高到91.7% ,精确率仍保持在90%以上,而理解错误率降低了13.5% ,同时计算复杂度的上升几乎可以忽略。
Abstract
In order to build a more accurate and robust voice command system , a novel Word Graph Expansion algorithm for voice command understanding is presented in this paper. It has been proved by experimental results that this algorithm has a much better performance than the generally adopted N-best algorithm while maintaining high computation efficiency. Also an error tolerance method is put forward to improve the robustness of our voice command understanding module ,which further decreases the understanding error rate (UER) to 16.6% with the computation efficiency almost unchanged compared with the case without error tolerance.
关键词
语音命令 /
N-best路径理解算法 /
词图扩展 /
自顶向下的图表句法分析方法 /
容错
{{custom_keyword}} /
Key words
voice command /
N-best paths understanding algorithm /
Word Graph Expansion /
Top-Down Chart parsing /
error tolerance
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Bernd Souvignier ,Bernhard Rueber ,et al. ,“The Thoughtful Elephant :Strategies for Spoken Dialog Systems”, IEEE Transactions on Speech and Audio Processing ,Vol. 8 ,No. 1 ,Journey ,2000. pp. 51 - 62
[2] Victor Zue ,Stephanie Seneff ,et al. ,“JUPITER:A Telephone-Based Conversational Interface for Weather Information”,IEEE Transactions on Speech and Audio Processing ,Vol.8 ,No.1 ,January ,2000. pp. 85 - 96
[3] Allen ,James. “Natural Language Understanding”,1993 ,Benjamin/Cummings Pub. Co. ,pp. 41 - 75
[4] Li Juanzi,“A Speech Understanding Model Based on Semantic Dependency Relationships”,ICCC’2001 ,Singapore.
[5] Hiroyuki Tsuboi and Yoichi Takebayashi ,“A Real-Time Task-Oriented Speech Understanding System Using Keyword-Spotting”, ICASSP’1992 ,Vol.1 ,1992 ,pp. 197 - 200
[6] Marc Hofmann and Manfred Lang ,“Intention - based Probabilistic Phrase Spotting for Speech Understanding”, Proceedings of 2001 International Symposium on Intelligent Multimedia ,Video and Speech Proceeding ,pp. 99 - 102
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家863高技术项目(863-306-ZD03-02-1);985重大项目“人机自然语言交互技术”(985校-22-攻关-06)
{{custom_fund}}
PDF(243 KB)
