Abstract
This study aims to propose a phonetic transcription and annotation system for automated speaking assessment of Korean. We analyzed the pronunciation errors discovered in the speech data of 2,250 sentences (32,610 words) read by 30 speakers of Chinese, Vietnamese, and Thai. We were able to predict the observed errors based on the comparative analyses between the phonetic and phonological system of Korean and those of the native languages of the learners, and thus suggested using the frequency of errors for the assessment of segmental accuracy. These results are expected to contribute to the on-going discussions of automated assessment of Korean speaking tests and the future development of automated tools for Korean language education.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
