Abstract
In recent years, music education has undergone profound shifts owing to the rapid advancements in digitalisation and the integration of big data technologies. Traditional methodologies in this domain have been challenged, particularly in the selection of musical tracks and their alignment with evolving pedagogical styles, thereby highlighting a pressing need for more contemporary, personalised, and data-driven approaches. This study delves into innovative methodologies designed to augment music education, primarily by harnessing big data technologies. Emphasis is placed on the extraction of salient features from musical tracks, including chroma, Mel Frequency Cepstrum Coefficient (MFCC), and Mel spectrogram. By conducting nuanced similarity analyses, recommendations for stylistic tracks were formulated. Furthermore, with the introduction of weighted calculations centred on musical significance, enhanced precision in track recommendations has been observed. Through these findings, this study not only deepens the academic discourse surrounding music education but also offers crucial insights for its practical applications.
Keywords
Get full access to this article
View all access options for this article.