DocumentCode :
2615261
Title :
Lattice Segmentation Based Confusion Network Generation Method
Author :
Wang, Huan-liang ; Zhu, Xi-jun ; Han, Ji-qing
Author_Institution :
Sch. of Inf. Sci. & Technol., Qingdao Univ. of Sci. & Technol., Qingdao, China
fYear :
2009
fDate :
17-20 April 2009
Firstpage :
141
Lastpage :
144
Abstract :
Confusion network (CN), a compact representation of lattice, has been arousing more and more attention in the area of speech recognition. In the paper, a novel lattice segmentation based CN generation method is proposed for remarkably reducing time complexity. Confidence based lattice segmentation algorithm is designed to cut lattices. The experimental results on a Chinese continuous speech database show that the significant reduction in generation time is achieved but almost no decline in the quality of CN compared to word-clustering algorithm (WCA) without arc-pruning operation. In the case of consuming same time, recognition performance of the proposed method is beyond that of WCA.
Keywords :
computational complexity; lattice theory; speech recognition; Chinese continuous speech database; arc-pruning operation; confidence based lattice segmentation algorithm; confusion network generation method; lattice cutting method; lattice representation; speech recognition; time complexity; word-clustering algorithm; Computer networks; Computer science; Computer science education; Databases; Decoding; Educational technology; Information science; Lattices; Natural languages; Speech recognition; confusion network; lattice segmentation; speech recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Education Technology and Computer, 2009. ICETC '09. International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-0-7695-3609-5
Type :
conf
DOI :
10.1109/ICETC.2009.75
Filename :
5169469
Link To Document :
بازگشت