Efficient keyword spotting system for information retrieval | |
Luo Jun ; Ou Zhi-jian | |
2010-05-06 ; 2010-05-06 | |
关键词 | Practical Theoretical or Mathematical Experimental/ audio signal processing filtering theory grammars graph theory information retrieval maximum likelihood estimation regression analysis speech recognition/ two-stage keyword spotting system audio information retrieval mass speech data searching stage preprocessing stage unsupervised MLLR adaptation maximum likelihood linear regression forward-backward algorithm filtering syllable N-grammar/ B6130E Speech recognition and synthesis B6140B Filtering methods in signal processing B0250 Combinatorial mathematics B0240Z Other topics in statistics C7250R Information retrieval techniques C1250C Speech recognition C5260S Speech processing techniques C1140Z Other topics in statistics |
中文摘要 | A new two-stage keyword spotting system was proposed based on syllable graph for audio information retrieval task, which could efficiently spot the interesting words in mass speech data, thus serve for the national security. It comprised two stages-preprocessing and searching. In the preprocessing stage, the audio data was recognized into syllabic graph which included high accuracy syllable candidates, and unsupervised MLLR (maximum likelihood linear regression) adaptation was carried out iteratively to further improve the accuracy of the syllable graph. In the searching stage, to answer the frequent queries from users, searching for matched keywords was only scanned in the graph for likely syllable strings. A forward-backward algorithm based on syllable N-grammar was used to calculate confidence measures for further filtering of the searching result. Experimental results show the system achieved good performances in both recall rate and accuracy rate, and in the searching stage only 0.01 times of real time is needed, which can meet the demand for fast retrieval. |
语种 | 中文 ; 中文 |
出版者 | People's Posts & Telecommun. Publishing House ; China |
内容类型 | 期刊论文 |
源URL | [http://hdl.handle.net/123456789/10853] ![]() |
专题 | 清华大学 |
推荐引用方式 GB/T 7714 | Luo Jun,Ou Zhi-jian. Efficient keyword spotting system for information retrieval[J],2010, 2010. |
APA | Luo Jun,&Ou Zhi-jian.(2010).Efficient keyword spotting system for information retrieval.. |
MLA | Luo Jun,et al."Efficient keyword spotting system for information retrieval".(2010). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论