The pedagogical potential of using neural networks in the field of professional music education

Introduction. Artificial intelligence is actively used in various spheres of life, including education. Recently, it has begun to be used in music education. This area is of interest, as it helps to improve the learning process and make it more accessible. It is necessary to ensure that these methods really help students and make the learning process more effective. The aim of this study: to conduct a systematic review of scientific papers that examine the pedagogical potential of artificial intelligence (neural networks) in the field of professional music education, implemented at the level of higher or secondary vocational education. Materials and methods. To achieve the stated objective of the study, a systematic literature review method was used. This review was carried out in accordance with the PRISMA 2020 protocol. The study used the Web of Science (WoS), Scopus, and Russian Science Citation Index (RSCI) databases. The final sample size was 31 publications (n = 31). Results. An increase in research interest in this issue has been established since 2022. The greatest activity of Chinese scientists has been revealed. Disproportions have been found between theoretical (19.4%) and empirical (80.6%) works. Six roles of using neural networks in music education have been identified: intelligent learning (48.6%), automated assessment systems (20%), blended learning (17.1%), personalized learning (8.5%), assistive technologies (2.9%), and virtual reality technologies (2.9%). The methods of teaching music (61.3%), control (54.8%), pedagogical communication with feedback (35.5%), assessment (25.8%), educational content (19.4%), and educational environment (16.1%) are most productively transferred to the field of artificial intelligence. The effects of using neural networks in music education include an increase in the quality of education (87%), academic performance (71%), the quality of musical performance in general (38.7%) or individual performance skills, such as musical expressiveness (19.3%), artistic literacy (6.4%), involvement in the educational process, and motivation for students to learn (45.2%). KEYWORDS Conclusions. Analysis of the practices of using neural networks in music education shows a deep gap between technological solutions and understanding of the pedagogical potential of these practices. Prospects for further research on this topic can contribute to the formation of digital didactics of music education.