QMUL, School of Electronic Engineering and Computer Science
Centre for Digital Music Seminar Series
Seminar by:
Johanna Devaney (Brooklyn College)
Date/time: Wednesday 21st of April, 4.30-5.30 pm
Location: Online
Open to students, staff, alumni, public; all welcome. Admission is FREE, no pre-booking required.
Title: Automatically Extracting Performance Data from Musical Audio
Abstract: This talk will discuss what we can observe about musical performance in the audio signal and where Music Information Retrieval (MIR) techniques have succeeded and failed in enhancing our understanding of musical performance. Building on my own work developing tools for analyzing musical performance, I will discuss the advantages and disadvantages of score-informed and blind approaches to extracting performance information in both monophonic and polyphonic audio. Video:
Bio: Johanna Devaney is an Assistant Professor at Brooklyn College and the CUNY Graduate Center, where she teaches courses in data analysis, music technology, music theory, and sonic arts. Her research focuses on interdisciplinary approaches to the study of musical performance, with a particular focus on the relationship between pitch structure and intonation in the singing voice. More broadly, she examines the ways in which recorded performances can be used to study and model performance and develops computational tools to facilitate this. Johanna's work draws on the disciplines of music, computer science, and psychology, and has been funded by the Social Sciences and Humanities Research Council of Canada (SSHRC), the Google Faculty Research Awards program and the National Endowment for the Humanities (NEH) Digital Humanities program.