DEVELOPMENT OF AI-BASED CLASBOT AS AN INNOVATIVE LEARNING MEDIA IN EAR TRAINING COURSES IN UNIMED MUSIC EDUCATION STUDY PROGRAM
Uyuni Widiastuti (a*), Herna Hirza (b), Adina Sastra Sembiring (c), Erizon (d)

Music Education Study Program, Faculty of Language and Arts, Universitas Negeri Medan

Abstract

This research is titled Development of Artificial Intelligence-Based Clasbot as an Innovative Learning Media in Ear Training Course at Unimed Music Education Study Program. The designed application, Clasbot, is specially customized to support learning in Ear Training course. The background of this study departs from the conventional method of learning Ear Training, which is by reading block notation and tapping rhythmic patterns available in the textbook. The purpose of this study is to design and develop an AI-based Clasbot as an innovative learning tool in Ear Training courses, and test the validity, practicality, and effectiveness of the Clasbot application. This research uses a Research and Development (R&D) approach by adopting the 4D model developed by Thiagarajan, which consists of four main stages: defining, designing, developing, and disseminating. Data collection was conducted through validity and reliability tests, practicality assessment, and effectiveness testing. Data analysis includes evaluating the validity of the instrument and testing the effectiveness of the application as a learning media. The final product is expected to encourage students to learn more independently and creatively. The results of the development of Clasbot as an innovative learning media show the following: (1) Assessment from material expert validators showed an average of 89.22%, categorized as very valid- (2) Assessment from media expert validators reached an average of 86.88%, also in the very valid category- (3) Practicality according to lecturers obtained an average score of 88.3%, indicating that this application is very practical- (4) Practicality according to students received an average score of 84.39%, which is also categorized as very practical- (5) Learning outcomes in the control class (class A) resulted in an average score of 31.28- (6) Learning outcomes in the experimental class (class B) reached an average of 32.08. From the effectiveness analysis of the Clasbot application, the statistical decision states that H&#8320- is rejected, which means that there is a significant difference between student learning outcomes in the control class and the experimental class. Thus, it can be concluded that Clasbot as an AI-based learning media is effective in improving student learning outcomes in Ear Training courses in the Unimed Music Education Study Program.

Keywords: Development- Clasbot- Media- Innovative- Ear Training

Topic: Multimedia and e-learning system