Skip to main content

Daniel Shanahan - "Charles Seeger's Tin Can: Machine Transcription

Monday, January 11, 2027 | 4:30 PM - 6:00 PM CT
University Hall, Hagstrum 201, 1897 Sheridan Road, Evanston, IL 60208 map it

Speaker

Daniel Shanahan, Northwestern University

Title

Charles Seeger's Tin Can: Machine Transcription and the Limits of Musical Measurement

Abstract

The extraction of data from musical performance is at the heart of our listening environment, with songs algorithmically recommended based on the measured similarity of factors such as "danceability" and "acousticness," alongside features such as tempo and key. The mechanical extraction of data from performance has been debated for nearly a century. The question is what it means to take a continuous stream of music and parse it into something a machine can read, and how best to do so in a way that aligns with what a listener might find most salient. Musicologist Charles Seeger set out to capture musical performance without the interference of a human transcriber through his "melograph," a device at first consisting of a cardiogram machine held together by a tin can found in a backyard. This talk approaches Seeger's device as an epistemic thing—an object that generated the very research questions it was built to answer. Drawing on recent archival research, it reconstructs what Seeger believed his machine could do and, just as importantly, what he acknowledged it could not. I argue that, while many of the compromises made in this early work persist in the field today, they are too often positioned as technological constraints rather than as the broader epistemological impasses they were initially understood to be.

Biography

Director of Graduate Music Studies. Daniel Shanahan’s research incorporates music-analytic, computational, and experimental methods to better understand the cognitive and communicative constraints of music. His interests include corpus studies, music and emotion, the oral transmission of music, the computational analysis of jazz and folk music, as well as machine learning models and generative artificial intelligence (AI).

Daniel’s work has been published in Music Perception, The Journal of New Music Research, Musicae Scientiae, The Journal of Jazz Studies, The Journal of the Acoustical Society of America, Cognition and Emotion, Frontiers in Psychology, and Empirical Musicology Review, among others. He contributed chapters on tonality, harmony, and counterpoint to The Routledge Companion to Music Cognition (Routledge, 2017), The Routledge Companion to Music Theory Pedagogy (Routledge, 2020), and has chapters in Over and Over Again: Exploring Repetition in Popular Music (Bloomsbury, 2018), as well as multiple chapters in the Oxford Handbook of Music and Corpus Studies (Oxford University Press, forthcoming), for which he is also serving as co-editor. He has contributed book reviews to Theory and Practice, and Music Theory Online.

Cost: FREE

Audience

  • Faculty/Staff
  • Student
  • Public
  • Post Docs/Docs
  • Graduate Students

Contact

Janet Hundrieser
(847) 491-3525
Email

Interest

  • Academic (general)
  • Arts/Humanities
  • Data Science & AI
  • Sciences
  • Social Sciences
  • Technology/Innovation

Add Event To My Group

Please sign-in