Exploring the Diverse Practices of Artificial Intelligence in Blended Music Teaching in Colleges and University

Abstract

Abstract With the development of the information age and the change in media technology, the education industry has also ushered in a new change, and this paper focuses on the application of artificial intelligence in blended music teaching in colleges and universities. The first part of the article utilizes artificial intelligence to appreciate music emotions, utilizing Thayer’s two-dimensional emotion model and emotion vectors to express emotions in music. Then, the short- and long-time feature extraction of music emotion is carried out separately and normalized. Then, a music emotion recognition model based on the CNN-SVM model is constructed by combining a deep learning network and a space vector machine model. The constructed music emotion recognition model and SPOC hybrid teaching are utilized to establish a blended music diversity curriculum in colleges and universities and practical exploration is conducted. The results show that the time range of the change in music emotion perception is about 2.5~3.2s, and the recognition success rate is more than 4/5 for all kinds of music emotion segments. The correlation between the effect of music teaching and music diversification practice and students’ satisfaction is less than 0.01. The diversification of music teaching in this study can effectively stimulate students’ learning interest and improve learning efficiency in the music classroom.