Digital Culture & AudioVisual Challenges

Summary

 This survey is part of a broader research initiative on machine learning (ML) software that we are developing for use in music education. With the ultimate goal of informing our software design, during the past academic year, we conducted a quantitative study, targeting music educators working in public music schools and conservatories in Greece.  Through this research, we examined teachers' habits and perceptions regarding the use of digital music technology, ML, and artificial intelligence (AI) in music education, as well as their views on the role of kinesthetic learning in the music classroom.  The analysis of the results yielded significant findings that help us piece together the needs of a highly kinesthetic and musically rich classroom—one that ML software could support and enhance.

1. Theoretical Framework

Although research on AI applications in music education is still in its early stages [1], it is evident that it has the potential to transform the field, making music education more innovative and efficient [2]. Relevant to this study, recent research highlights the transformative potential of AI in education, offering increased accessibility for diverse learners [3] and enhanced engagement through interactive tools [4]. Technologies such as gesture recognition, music information retrieval, and motion capture enable innovative, movement-based teaching methods [5].

2. Research objectives

 Our pursuit stems from the idea of putting movement-based teaching into the music class to enhance learners' high-level skills using ML tools. Our under-development software [6] and user experiences aim to support instructors and students by adding kinesthetic features to the learning process. This study explores how ML can aid kinesthetic learning, focusing on (a) challenges in adopting AI tools and (b) the potential pedagogical benefits and value of our support tools in teaching practice.

3. Methodological considerations

 To fulfill our objectives, we targeted public music school and conservatory educators in Greece, a total of 114. As the aim was the collection of a broader sample of music educators, we contacted professionals from all scattered parts of Greece, employing a quantitative survey approach, collecting data via an online questionnaire. Our design covered disparate theme sections including demographics, the use of AI, perceptions of AI in kinesthetic learning and their willingness to adopt AI-based Tools.  

4. Results and discussion

The survey results indicate limited current use of ML and AI in music education, but teachers show interest in adoption. Most have computer access but need additional equipment and support. A majority are willing to try AI-driven kinesthetic tools.

The analysis provided valuable insights, allowing us to better understand the requirements of a classroom that emphasizes kinesthetic learning and musical engagement. The trade-offs of introducing an AI-based highly kinesthetic teaching model rely on teachers’ concerns about the impact of AI tools on students’ skills and in the challenges they face in their daily teaching environments.

5. Conclusion

 The software's perceived value includes promoting inclusion through kinesthetic methods, simplifying musical activities, and increasing participation. Teachers recognize technology as beneficial, and ML could help address instrument shortages and student engagement. Potential pedagogical benefits include improved student involvement and understanding of musical concepts, particularly through interactive applications.

Nevertheless, the necessary conditions for new teaching models to significantly enhance learning are set out clearly. This leads us to a set of design rules for our software to comply to.

 References

1. Zhang, B. W. Fen, C. Zhang, and S. Pi, “Transforming Music Education Through Artificial Intelligence: A Systematic Literature Review on Enhancing Music Teaching and Learning.” International Journal of Interactive Mobile Technologies, vol. 18, no. 18, 2024.
2. Shang, “The application of artificial intelligence in music education,” in Intelligent Computing Theories and Application: 15th International Conference, ICIC 2019, Nanchang, China, August 3–6, 2019, Proceedings, Part II 15. Springer, 2019, pp. 662–668. 
3. Tremblay, Pierre Alexandre, Owen Green, Gerard Roma, James Bradbury, Theodore Moore, Jacob Hart, and Alex Harker. "Fluid corpus manipulation toolbox." (2022).
4. Fiebrink, R., & Cook, P. R. (2010, August). The Wekinator: a system for real-time, interactive machine learning in music. InProceedings of The Eleventh International Society for Music Information Retrieval Conference (ISMIR 2010) (Utrecht) (Vol. 3, pp. 2-1).
5. Jourdan, T., & Caramiaux, B. (2023, May). Machine learning for musical expression: A systematic literature review. InNew Interfaces for Musical Expression (NIME).
6. Smailis, D., & Heliades, G. P. (2023). Elaborating Advanced Machine Learning Techniques in the Music Class.European Journal of Engineering and Technology Research, 107-113.