Statistical Analysis of Korean Pronunciation Variations

In this paper, we present a statistical analysis of Korean pronunciation variations using a Grapheme-to-Phoneme (GTP) system. The GPT system generates pronunciation variants by applying rules modeling obligatory and optional phonemic changes and allophonic changes in spoken Korean. Experimental results using a PBS (Phonetically Balanced Sentence) Speech DB of 60,000 sentences show that the most frequently happening obligatory phonemic variations are in the order of liaison, tensification, aspirationalization, and nasalization of obstruent, and the most frequently happening optional phonemic variations are in the order of initial consonant /h/-deletion, insertion of final consonant with the same place of articulation as the next consonant’s, and deletion of final consonant with the same place of articulation as the next consonant’s. These statistics can be used for improving the performance of speech recognition systems.