Graduate Inst. of Commun. Eng., National Taiwan Univ.Tsai, Ming-yiMing-yiTsaiChou, Fu-chiangFu-chiangChouLIN-SHAN LEE2007-04-192018-07-052007-04-192018-07-052001-12N/Ahttp://ntur.lib.ntu.edu.tw//handle/246246/2007041910043003https://www.scopus.com/inward/record.uri?eid=2-s2.0-84962892877&doi=10.1109%2fASRU.2001.1034587&partnerID=40&md5=d75f97e574965c3087f9d878c5ba9af7We propose a new approach to rank the potential pronunciations for each word by their pronunciation frequency and inverse word frequency (pf-iwf) weights. The pronunciation set obtained in this way can then be pruned with different criteria. This approach not only considers the frequencies of occurrence of the pronunciations, but tries to minimize the extra confusion which may be introduced by pronunciation variations, such that the best overall performance can be achieved. A new entropy-based approach for pruning the pronunciation variations is also proposed. Experimental results showed that the proposed approach can not only improve the recognition performance, but make the performance more stable and less sensitive to various parameters, factors and options including the different pruning criteria. All the experiments were performed with the LDC Mandarin Call Home corpus, although the approaches and principles are definitely not limited to Mandarin Chinese. © 2001 IEEE.application/pdf378240 bytesapplication/pdfen-US[SDGs]SDG4Data mining; Entropy; Linguistics; Entropy based approach; Mandarin Chinese; New approaches; Pronunciation modelling; Pronunciation variation; Word frequencies; Speech recognitionconference paper10.1109/ASRU.2001.10345872-s2.0-84962892877http://ntur.lib.ntu.edu.tw/bitstream/246246/2007041910043003/1/01034587.pdf