부산대학교 도서관

With the rapid development of artificial intelligence technology and its deep integration with education, the use of intelligent means to analyze and study teaching behaviors has become a research hotspot. Teaching behavior can be divided into verbal behavior and non-verbal behavior, among which verbal behavior is the main way of classroom teacher-student interaction, accounting for about 80% of all teaching behaviors. At present, teaching researchers mainly use manual annotation to code and analyze teaching verbal behaviors, which has the problems of low efficiency and individual subjectivity, and it is difficult to collect a large amount of teaching verbal behavior data to analyze teaching patterns. This paper designs and implements a teaching verbal behavior analysis tool using teaching verbal behavior recognition algorithm. It provides the functions of speech transcription of teaching videos, automatic recognition and analysis of teaching verbal behaviors, and analysis of teaching verbal contents, with a view to providing assistance to teaching researchers in analyzing teaching verbal behaviors in instructional videos by using intelligent technology.