ScholarWorks@Y-Scholar Hub

초록

This paper explores the direction of Korean learners’ spoken corpus, focusing on the problems within the transcription and distribution process of spoken corpus. This paper points out that inter-transcriber reliability poses a problem in the process of transcribing the learners’ spoken interlanguage. To solve this problem, this paper proposes a method of displaying the locations of pronunciation errors while maintaining transcription at an orthographic level under the assumption that the audio files are disclosed. For the discussion of the distribution process, this paper refers to the learners’ corpus abroad. This study confirms that the end-user should be considered in the distribution process as spoken corpus is used by human. In relation to this, the provision of a search service, analysis tools, and user manuals are proposed as a specific plan. In addition, this paper suggests that speaker classification and search by genre are necessary improvements in the distribution of the Korean learners’ spoken corpus currently provided by the National Institute of the Korean Language. This paper is meaningful in that it suggests a direction for constructing a Korean learners’ spoken corpus, which has been lacking in the past.