International Symposium on Chinese Spoken Language Processing (ISCSLP 2002)

Taipei, Taiwan
August 23-24, 2002

Linguistic and Acoustic Analysis of Chinese Person Names

Wen-Jie Cao, Bo Xu, Juha Iso-Sipila

(1) National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences; (2) Nokia China R&D Center, Beijing, China 112

In this paper, we give results on our recent study on Chinese person names. The analysis is based on a corpus of 1 million names. The results include the syllable lengths and surname composition of the names in the corpus, full name and given name statistic results and analysis, tonal pattern analysis of Chinese full and given names, and name confusion analysis when given number of names are extracted from the corpus. Since pronunciation is involved in the tonal pattern of the names and confusion analysis, we give our strategy of dealing with the multi-pronunciation characters in Chinese person names. A model is brought forward to estimate the upper-limit to the recognition accuracy when N words are randomly extracted from the corpus, and has been used in our name analysis.


Full Paper

Bibliographic reference.  Cao, Wen-Jie / Xu, Bo / Iso-Sipila, Juha (2002): "Linguistic and acoustic analysis of Chinese person names", In ISCSLP 2002, paper 103.