Entering Tone Recognition in a Support Vector Machine Approach

Author

Wang, Xiangcheng ; Liu, Ying ; Cai, Lianhong

Author_Institution

Dept. of Chinese Language & Literature, Tsinghua Univ., Beijing

Volume

2

fYear

2008

fDate

18-20 Oct. 2008

Firstpage

61

Lastpage

65

Abstract

This paper introduces support vector machine classifiers into entering tone recognition. Not every syllable needs recognition in a statistical way. The recognition accuracy of syllables which need recognition in the support vector machine approach is about 90%, which makes it possible to analyze poems´ rhymes and translate Mandarin into many Chinese dialects. The experiments also check the influence of context window attributes on the entering tone recognition.

Keywords

learning (artificial intelligence); natural languages; signal classification; speech recognition; statistical analysis; support vector machines; Chinese dialect; Mandarin translation; context window attribute; entering tone recognition; poem rhyme analysis; statistical learning; support vector machine classifier approach; syllable recognition; Asia; Computer science; Handwriting recognition; Kernel; Natural languages; Polynomials; Statistical learning; Support vector machine classification; Support vector machines; Tagging; Support Vector Machine; entering tone; pinyin;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2008. ICNC '08. Fourth International Conference on

Conference_Location

Jinan

Print_ISBN

978-0-7695-3304-9

Type

conf

DOI

10.1109/ICNC.2008.757

Filename

4666957