메뉴 건너뛰기
.. 내서재 .. 알림
소속 기관/학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
로그인 회원가입 고객센터 ENG
주제분류

추천
검색
질문

논문 기본 정보

자료유형
학술대회자료
저자정보
Cellik Adams (Sogang University) Hyukche Kwon (Sogang University) EunKyoung Jo (Sogang University) Sihun Lee (Sogang University) Myoung-Wan Koo (Sogang University)
저널정보
한국HCI학회 한국HCI학회 학술대회 PROCEEDINGS OF HCI KOREA 2025 학술대회 발표 논문집
발행연도
2025.2
수록면
42 - 48 (7page)

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
🏆
연구결과
AI에게 요청하기
추천
검색
질문

초록· 키워드

오류제보하기
In the field of AI-assisted education, the development of high-quality, domain-specific datasets remains a challenge, particularly for languages beyond English. This paper presents a novel multi-stage pipeline for processing Korean tutoring dialogues, addressing the need for accurate transcription and speaker labeling of educational interactions. Our approach combines state-of-the-art automatic speech recognition (Whisper), speaker diarization (pyannote.audio), and large language model synthesis (Claude) to effectively process educational audio recordings. Our pipeline successfully handles the complexities of Korean educational discourse, including natural code-switching behaviors and the cultural context necessary for accurate speaker identification. We detail our methodology for resolving conflicts between whole-file and segment-wise transcriptions, as well as our approach to speaker role identification using contextual analysis. The pipeline demonstrates effective handling of Korean educational discourse, including natural code-switching behaviors, while maintaining high transcription quality and accurate speaker attribution. This work establishes reproducible methods for creating similar resources in other educational contexts, contributing to the broader goal of developing sophisticated AI-assisted educational systems.

목차

Abstract
1. Introduction
2. Background
3. Methodology
4. Results and Discussion
5. Limitations and Future Work
6. Conclusion
References

참고문헌 (0)

참고문헌 신청

함께 읽어보면 좋을 논문

논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!

최근 본 자료

전체보기

댓글(0)

0