Geometric Analysis of Musical Audio Data

Project Summary

In this work, we turn musical audio time series data into shapes for various tasks in music matching and musical structure understanding. 

Themes and Categories
Year
Contact
Christopher Tralie
Electrical and Computer Engineering
chris.tralie@gmail.com

In particular, we use sliding window representations of chunks of audio to create high dimensional time-ordered point clouds, and we extract information by analyzing the geometry of these clouds.  We have shown, for example, that sequences of these shapes can be used to identify two different versions of the same song, or "cover songs."  We have also shown that both local and global musical properties can be expressed in geometric language.  For instance, hip hop is very "wiggly" while classical music is very "smooth."  Choruses and verses tend to live in distinct clusters connected by paths, and "bridges," or detours to a different musical idea, show up as large loops.

Related People

Related Projects

A team of students led by Co-Principal Investigators Dr Jenny Immich and Dr Vicky McAlister will develop a geospatial methodology to automate data analysis originating from small unmanned aerial vehicles (SUAV) that seeks to identify the homes of ordinary medieval people within the modern Irish landscape. Known as aerial archeology, this work investigates archeological remnants on the surface of the earth without excavation. Students will create iterative Python scripts to analyze elevation datasets (hillshades) in areas surrounding medieval castles. Students will create advanced analysis methodologies for exploring low-lying archeological features in the landscape, including principal component, standard deviation, and difference mean elevations. Students with interest and appropriate backgrounds will be challenged to develop machine learning methods for identification of linear archeology. The resulting portfolio of work will be robust, including workflows, metadata, methodologies, and scripts. This work will solidify geospatial methodologies in advancing our knowledge of where everyday people lived in Ireland during the medieval period, a question that has long perplexed archeologists. In addition, it will fundamentally ground our understanding of complex topics in urbanization to the modern period and shed light on the development of communities across time. 

 

Project Lead: Vicky McAlister, Southeast Missouri State University

Project Manager: Harshit Sahay 

Is there a right type and amount of consumption? The idea of ethical consumption has gained prominence in recent discourse, both in terms of what we purchase (from fair trade coffee to carbon off-sets) and how much we consume (from rechargeable batteries to energy efficient homes). These modes of ethical consumerism assume that individuals become political, as well as economic, actors through shopping. Concern with the morality of consumption is not new to capitalist societies, and we can see as much in the earliest discourses surrounding the market economy. During the Middle Ages and the Renaissance, acts of consumption became increasingly aligned with corruption, as individual and corporate bodies were depicted as altered, even damaged, by trade.  And, in an increasingly global and colonial economy, authors debated the ethics of expanding into new markets from India to the Americas. These questions and others like them have shaped our own modern discourse around the ethics of consumption.  

 

Building on the past two years of work, this project will extend the analysis of consumption performed in 2020 and 2021 to include the data collected and analyzed by the Bass Connections team from the rare materials archives of Duke’s Rubenstein library and of the University of Alabama Birmingham’s Lister Hill library. These materials include Medieval and Early Modern medical manuals, which will help us to analyze the metaphorical translation from medical consumption of the human body to economic consumption of the body politic; seventeenth-century mercantile theory texts, which theorize how consumption is connected to trade and monetary policy; as well as treatises and legislative documents on monopolies, corporations, and trading companies. 

 

Project Leads: Astrid Giugni, Jessica Hines

Project Manager: Erin Winter

This two-week teaching module in an introductory-level undergraduate course invites students to explore the power of Twitter in shaping public discourse. The project supplements the close-reading methods that are central to the humanities with large-scale social media analysis. This exercise challenges students to consider how applying visualization techniques to a dataset too vast for manual apprehension might enable them to identify for granular inspection smaller subsets of data and individual tweets—as well as to determine what factors do not lend themselves to close-reading at all. Employing an original dataset of almost one million tweets focused on the contested 2018 Florida midterm elections, students develop skills in using visualization software, generating research questions, and creating novel visualizations to answer those questions. They then evaluate and compare the affordances of large-scale data analytics with investigation of individual tweets, and draw on their findings to debate the role of social media in shaping public conversations surrounding major national events. This project was developed as a collaboration among the English Department (Emma Davenport and Astrid Giugni), Math Department (Hubert Bray), Duke University Library (Eric Monson), and Trinity Technology Services (Brian Norberg).