Data & Analytics

U Michigan Researchers Turn to Data Science to Understand Music

• By Joshua Bolkan
• 04/24/18

Four research teams at the University of Michigan will explore the intersection of music and data science thanks to the support of the Michigan Institute for Data Science's (MIDAS) Data Science for Music Challenge Initiative.

The challenge asked participants to propose research projects that applied data science tools such as data mining or machine learning to the study of areas such as music theory, the connection between music and words, performance and more. Possible areas of research suggested by the challenge's coordinators include algorithms and computer composition, big data-based instrument deign, music education method analysis, collaborative music making and music recommendation systems, among others.

The selected projects, each of which will receive a one-year $75,000 grant, include:

• Understanding and Mining Patterns of Audience Engagement and Creative Collaboration in Large-Scale Crowdsourced Music Performances, a project that aims to develop a platform for crowdsourcing music creation and performance and applies data-mining tools to discovering patterns related to audience participation and engagement. The researchers expect their findings to be relevant to other interactive settings, according to information released by U-M, such as in the development of educational tools;
• Understanding How the Brain Processes Music Through the Bach Trio Sonatas, which will use novel algorithms to analyze digitized performances of Bach's Trio Sonatas to determine what makes a performance artistic and where performers are likely to make mistakes. Researchers plan to integrate their findings into courses on brain performance and data science;
• The Sound of Text, which "will develop a data science framework that will connect language and music, developing tools that can produce musical interpretations of texts based on content and emotion. The resulting tool will be able to translate any text — poetry, prose or even research papers — into music," according to information released by the university; and
• A Computational Study of Patterned Melodic Structures Across Musical Cultures, a project that will compare melodies from a half-dozen cultures to find cross-cultural musical similarities.

"MIDAS is excited to catalyze innovative, interdisciplinary research at the intersection of data science and music," said Alfred Hero, co-director of MIDAS and the John H. Holland Distinguished University Professor of Electrical Engineering and Computer Science, in a prepared statement. "The four proposals selected will apply and demonstrate some of the most powerful state-of-the-art machine learning and data mining methods to empirical music theory, automated musical accompaniment of text and data-driven analysis of music performance."