DocumentCode :
3150529
Title :
Research on Chinese phonetic string segmentation of sentential input
Author :
Gao, Yudong ; He, Jiaming ; Li, Jinxiang
Author_Institution :
Fac. of Inf. Sci. & Eng., Ningbo Univ., Ningbo, China
fYear :
2011
fDate :
16-18 April 2011
Firstpage :
4334
Lastpage :
4337
Abstract :
Syncopation of Chinese phonetic string is the basis of Chinese Sentential Input. The traditional matching algorithm of phonetic string only brings one segmentation result and thus leads to the incomplete containing of the ambiguous results and the failure of the phonetic string conversion. The paper raised a improved method for the syncopation of Chinese phonetic string. By adopting a Bidirectional Maximum Matching algorithm and combining the syncopation rules of double vowels, we are able to preserve all the disambiguity possibilities during the phonetic syncopation which would participate in the competition for the optimal sentence candidate, so that we can ensure the correctness of the conversion of Chinese phonetic string.
Keywords :
pattern matching; speech processing; Chinese phonetic string segmentation; Chinese phonetic string syncopation; bidirectional maximum matching algorithm; double vowel syncopation rules; sentential input segmentation; Algorithm design and analysis; Computers; Error analysis; Helium; Industries; Law; Bidirectional Maximum Matching; Chinese phonetic string; Maximum Matching Method; Reverse Maximum Matcing Method;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Consumer Electronics, Communications and Networks (CECNet), 2011 International Conference on
Conference_Location :
XianNing
Print_ISBN :
978-1-61284-458-9
Type :
conf
DOI :
10.1109/CECNET.2011.5768344
Filename :
5768344
Link To Document :
بازگشت