MusicCritic: A technological framework to support online music teaching for large audiences

TitleMusicCritic: A technological framework to support online music teaching for large audiences
Publication TypeConference Paper
Year of Publication2018
Conference Name33rd World Conference of International Society for Music Education (ISME)
AuthorsBozkurt, B., Gulati S., Romani O., & Serra X.
Conference Start Date15/07/2018
Conference LocationBaku, Azerbaijan
AbstractThis paper concerns online music education and as contribution, it proposes a new technological framework to support online music performance teaching to reduce loads on teachers for assessing large number of student performances. The online education field is growing exponentially. One form of online education is the Massive Open Online Courses (MOOCs) where large number of students, on the order of thousands, are enrolled to online courses. Recently, there have been course offerings for teaching music performance through MOOCs which basically rely on peer evaluation for the assessment of student performances and providing feedback. MOOCs designed for other domains such as computer programming have been successfully using supporting technologies that facilitate assessment and feedback. Here, we argue that supporting technologies dedicated to reducing instructor load in teaching music performance online would pave the way for successful MOOCs in this domain and provide new opportunities for music educators to reach larger audiences. In this paper, we propose a framework (MusicCritic, https://musiccritic.upf.edu ) that can help scale practice-based online music education upto MOOCs level without relying on peer evaluation methods. We discuss two main components of the framework. First, we consider the interfaces for setting up practice exercises, recording student performances, assessing the performances and providing feedback to the students. Second, tools for facilitating assessment are discussed where we demonstrate a semi-automatic assessment system that can learn from assessment of the instructor on a small group of performances and further assess larger sets of performances. We finally present tests performed on real-life data to demonstrate the potential of the approach.
preprint/postprint documenthttps://doi.org/10.5281/zenodo.1211450
intranet