Digital Musicology

“Applied computational and informatics methods for enhancing musicology”

Convenor: Dr Kevin Page

Hashtag: #digitalmusicology and #DHOxSS2019

Computers: Participants are not required to bring their own laptops for this workshop. Laptop computers will be provided by DHOxSS

 

Abstract

 

A wealth of music and music-related information is now available digitally, offering tantalizing possibilities for digital musicologies. These resources include large collections of audio and scores, bibliographic and biographic data, and performance ephemera -- not to mention the ‘hidden’ existence of these in other digital content. With such large and wide ranging opportunities come new challenges in methods, principally in adapting technological solutions to assist musicologists in identifying, studying, and disseminating scholarly insights from amongst this ‘data deluge’.

 

This workshop provides an introduction to computational and informatics methods that can be, and have been, successfully applied to musicology. Many of these techniques have their foundations in computer science, library and information science, mathematics and most recently Music Information Retrieval (MIR); sessions are delivered by expert practitioners from these fields and presented in the context of their collaborations with musicologists, and by musicologists relating their experiences of these multidisciplinary investigations.

 

Intended outcomes

 

The workshop comprises a series of lectures and hands-on sessions, supplemented with reports from musicology research exemplars. Theoretical lectures are paired with practical sessions in which attendees are guided through their own exploration of the topics and tools covered. The course covers computational approaches to three broad types of digital music data: audio data, symbolic data, and metadata. Delegates will learn the basics of each, their practical limitations, and understand which branches of musicology each is directly applicable to. Attendees will also gain an appreciation of how digital data and findings from musicologies other than their own can be selectively reused and repurposed, beyond original conception and intent, when structured approaches to data collection and processing are followed. Delegates are expected to leave the workshop with sufficient familiarity to select which digital musicology approaches are applicable to their own field of research; and equipped with sufficient insight, context, and confidence to seek out greater detail as an independent researcher.

Experience necessary

Familiarity with core music and musicological concepts and approaches is assumed. No prior technical knowledge is necessary. The course will introduce a wide variety of technical tools and methods in hands-on practical sessions, including editing music encodings and some introductory programming.

Convenor

Kevin Page is a senior researcher and associate member of faculty at the University of Oxford e-Research Centre, where he applies Linked Data to the Digital Humanities. He is investigator of the AHRC ‘Unlocking Musicology’ project, a co-investigator of ‘Digital Delius’, ‘Mapping Manuscript Migrations’ and ‘Workset Creation for Scholarly Analysis’, and runs the AHRC Linked Art research network. As Technical Director of Oxford Linked Open Data (OXLOD) he works with collections across the Gardens, Libraries, and Museums of the University, and has participated in W3C activities including the Linked Data Platform (LDP) working group. From 2012-15 he convened the Linked Data workshop at DHOxSS, where he now runs the Digital Musicology course.

Other speaker biographies

"Excellent teaching, and enormous patience and willingness to stop, go back and explain again."

DHOxSS 2018 participant

TIMETABLE

 
The Digital Musicology workshop will be held in one of the Sloane Robinson seminar rooms.
 
Link to overview of the week's timetable including evening events.
 
Monday 22nd July
08:00-09:00
Registration (Sloane Robinson building)
Tea and coffee (ARCO building)
09:00-10:00

Opening Keynote (Sloane Robinson O'Reilly lecture theatre)
10:00-10:30

Refreshment break (ARCO building)
10:30-12:00

Overview of the Digital Musicology
Over the coming week the Digital Musicology workshop at DHOxSS 2019 will introduce a wide variety of practical and theoretical digital techniques and illustrate their use within a number of musicology studies. Of course, in one week we can only scratch the surface of a myriad of methods and investigations - this talk contextualises the forthcoming lectures as representative of wider study, setting the scene within the wider digital musicology landscape.
Speakers: Kevin Page
Digital Musicology: a personal perspective/ An Introduction to Music Information Retrieval: musicological implications 
Computational tools, such as those of music information retrieval (MIR), are being enhanced and adapted to the needs of musicologists. These offer new, rapid and effective ways of investigating large collections of music in audio or score form. Recent advances in Web technology also allow researchers to record and share their results and working methods in a sustainable way, so their methods can be easily altered in the light of new knowledge or re-used on new data. These introductory lectures offer two personal distinct but complementary views from leading researchers in their fields on how these technological innovations can be brought to bear within musicology.
Speakers: David Lewis, Stephen Downie
12:00-13.30

Lunch (Dining Hall)
13:30-15:30

Hands on: Using computers to analyse recordings An introduction to signal processing
This session, and the following hands-on session, introduces the basics of computational treatment of recordings of music, which are based on the concept of ‘features’ derivable from this ‘signal’ by suitable processing. The hands-on session will expose you to software for extracting features from recordings, visualising those features, and will help you understand how features relate to perceptual and musical concepts.
Speakers: Stephen Downie, Chris Cannam
With: David Lewis, Kevin Page
15.30-16.00

Refreshment break (ARCO building)
16:00-17:00

Using computer analyses to index and find recordings Feature search and retrieval
Having previously covered the extraction of features from musical recordings, in this session you will be introduced to the technique of using geometrical distance to quantify the similarity between sets of features, and we will relate application of that technique to the task of finding recordings of interest within a larger collection.
Speaker: David Lewis

 

Tuesday 23rd July

 

09:00-10:30
Symbolic Music Analysis of Renaissance Counterpoint: current challenges

 

Speaker: Frauke Jurgensen

The Baudelaire Song Project: Digital Analysis of Song Settings

This session will present the digital tools used by the Baudelaire Song Project to analyse song settings of Charles Baudelaire’s poetry. With a particular emphasis on Sonic Visualiser, the project team will show how they use digital humanities methodologies to understand the interaction between words and music and the changes wrought upon poetic texts when they are set to music.

Helen Abbott, Caroline Ardrey, Nina Rolland

10:30-11:00

Refreshment break (ARCO building)
11:00-13:00

 

Training computers automatically to recognise patterns in recordings (Practical machine learning)
When analysing a large corpus of audio, a limiting factor is time: it is not practical to find patterns in a very large collections of audio by just listening. So-called ’machine learning’ techniques offer a means around this limit. In this session we will show you how to use modern machine learning techniques to distill out patterns in large collections of audio, without exhaustive human audition. A hands-on session on the same topic follows. 

Speakers: David Lewis, Stephen Downie

13:00-14:30

Lunch (Dining Hall)
14:30-15:30

Methods for analysing large-scale resources and big music data 
This session will take the tools of the last few sessions and consider the effects and consequences of scale. We will look at how you can manage and mitigate the problems of working with very large amounts of data. We will explore techniques that work best at this scale, using music collections from the British Library to define explore, analyse and compare large datasets across historic, cultural, and musical dimensions.
Speaker: Tillman Weyde
15:30-16:00

Refreshment break (ARCO building)
16:00-17:00
Lectures (various venues)

Wednesday 24th July
09:00-10:30

Digitised Notated Music: hands on with MEI 

There are two broad domains of digitised music: audio and so-called symbolic, which includes encodings of music notation. In this session we introduce the MEI music notation formats. We learn the models of music notation they employ, the text critical apparatus they provide, and how to prepare documents in these formats.

Speaker: David Lewis

With: Andrew Hankinson

10:30-11:00

Refreshment break (ARCO Building)
11:00-13:00

 

Digitised Notated Music: hands on with MEI (continued)
 
Annotating and structuring musicology knowledge using Linked Data 

The Semantic Web can be thought of as an extension of the WWW in which sufficient meaning is captured and encoded such that computers can automatically match, retrieve, and link resources across the internet that are related to each other. In a scholarly context this offers significant opportunities for publishing, referencing, and re-using digital research output. In this session we introduce the principles and technologies behind this ‘Linked Data’, illustrated through examples from musicological study.

Speaker: Kevin Page

13:00-14:30

Lunch (Dining Hall)
14:30-15:30

Annotating and structuring musicology knowledge using Linked Data (continued) 
15:30-16:00

Refreshment break (ARCO building)
16:00-17:00
Lectures (various venues)
Thursday 25th July
09:00-10:30
Automatic transcription of scanned notation: state of the art and applications; hands on with Gamera 
Speaker: Andrew Hankinson

10:30-11:00

Refreshment break (ARCO building)
11:00-13:00
Computer processing of digital notated music: hands on with music21 Working with symbolic music data
Given a corpus of digital musical documents, how can we explore its contents? In this session we introduce the music21 toolkit which allows us to search for patterns in such music corpora and to prepare reproducible analytic tools. We learn its specialist query language and some basic Python programming techniques.
Speaker: David Lewis
13:00-14:30

Lunch (Dining Hall)
14:30-15:30
Computer processing of digital notated music: hands on with music21 and programming in Python (continued)
Speaker: David Lewis 
15:30-16:00

Refreshment break (ARCO building)
16:00-17:00
Lectures (various venues)
Friday 26th July
09:00-10:30
An overview of software and data management best practice 
Revision control refers to a set of practices to track and control changes to your project files. Learn how to manage, revise, and collaborate on digital documents; how to revert files back to a previous state; and how to see when a particular change was introduced, and who was responsible.
Speaker: David Lewis
A case study in Early Music, from digitisation to musicological research
We base our presentation on our experiences with a large collection of images of historical music prints (EMO). Using optical recognition methods, we encoded a representative test-set automatically. We shall describe what further work is needed to enable useful and interesting searches, comparisons and other musical investigations on incorporating both the resulting corpus and relevant external resources.
Speaker: Tim Crawford
10:30-11:00

Refreshment break (ARCO building)
11:00-13:00
Hands on: from digitisation to analysis, an end-to-end example  
Speakers: Tim Crawford, David Lewis, Andrew Hankinson, Kevin Page
13:00-14:30

Lunch (Dining Hall)
14:30-15:30
Round table discussion: applied digital musicology in your research  
Speakers: Tim Crawford, David Lewis, Andrew Hankinson, Kevin Page, Stephen Downie

15:30-16:00

Refreshment break (ARCO building)
16:00-17:00
Closing plenary (Sloane Robinson O'Reilly lecture theatre)
Speaker Biographies

Helen Abbott is a Professor of Modern Languages at the University of Birmingham. She studied at the University of Cambridge and King’s College London, completing a PhD thesis on Baudelaire and Mallarmé. She specialises in nineteenth-century French poetry and music, exploring ways of writing about word-music relationships in poetic language, in critical theories, and using digital methodologies. She has published three books on Baudelaire, including the latest Baudelaire in Song

1880-1930 (Oxford University Press, 2017). Her particular focus is the work of (post-) romantic and symbolist poets including Gautier, Baudelaire, Verlaine, Rimbaud, Villiers de l’Isle-Adam, and Mallarmé. A classically-trained soprano, she connects her intellectual and personal interests in order to lead an international team of researchers on the Baudelaire Song Project (2015-2019, AHRC-funded).

Caroline Ardrey is Lecturer in French at the University of Birmingham and, since September 2017, has been Senior Research Associate on the Baudelaire Song Project (AHRC-funded 2015-2019). Having completed a DPhil on Stéphane Mallarmé's La Dernière Mode in 2014, Caroline joined the Baudelaire Song Project team in August 2015 as Research Associate, working on the initial phases of developing the Project's interdisciplinary dataset and digital song analysis techniques. Her wider research focuses on nineteenth-century French poetry and its interactions with popular culture and with other art forms, including music and fashion. Caroline is currently preparing a monograph, to be published with Routledge, which examines the reception of Baudelaire's poetry through popular song settings of his work.

Chris Cannam is Principal Research Software Developer in the Centre for Digital Music at Queen Mary University of London, where he works with researchers to produce useful software for music analysis. He is the primary author of the Sonic Visualiser application and many of its plugins. ​

Tim Crawford worked as a professional lutenist, playing on several recordings made during the 1980s. As a musicologist he studies lute music of the 16th to 18th centuries. Since the early 1990s he has been active in the rapidly-expanding field of MIR and was President of ISMIR for two years. He was PI of the AHRC-funded Transforming Musicology project, which was the immediate catalyst for the Digital Musicology workshops in DHOxSS.

J. Stephen Downie is a professor and the associate dean for research at the Graduate School of Library and Information Science, University of Illinois. Dr. Downie conducts research in music information retrieval. He was instrumental in founding both the International Society for Music Information Retrieval and the Music Information Retrieval Evaluation eXchange.  Downie is also the Illinois co-director of the HathiTrust Research Center which provides analytic access to the HathiTrust's

massive collections of digitized texts.

Andrew Hankinson is a Senior Software Engineer with the Bodleian Libraries in the Digital Research area. He earned a Masters in Library and Information Studies and a PhD in Music Information Retrieval from McGill University. Andrew specializes in working with large collections of digitized music documents, opening access to these collections by

using document image recognition technologies to make them searchable. He is also a member of the board of the Music Encoding Initiative, and a collaborator on the Single Interface for Music Score Searching and Analysis (SIMSSA) project.

Frauke Jürgensen (University of Aberdeen) Frauke's research interests include medieval and Renaissance compositional and performance practice, and the development of computer tools for the analysis of Renaissance counterpoint. From 2012 to 2014, in collaboration with McGill University and MIT, she was the UK PI for the “Electronic Locator of Vertical Interval Successions” (ELVIS) grant, funded through the “Digging into Data” programme. She has studied topics such as musica ficta and signature accidentals in the 15^th -century using symbolic music analysis, and is also active as a performer.

David Lewis is a researcher based at the Oxford e-Research Centre and the Royal Birmingham Conservatoire. He has recently worked on projects at Goldsmiths, University of London, Universität des Saarlandes and Universiteit Utrecht. His research focusses on the creation, dissemination and use of digital corpora of music (such as the Electronic Corpus of Lute Music) and music theory (earlymusictheory.org and Thesaurus Musicarum Italicarum).

Nina Rolland is a Research Associate in French studies at the University of Birmingham, working on the Baudelaire Song Project led by Professor Helen Abbott (AHRC-funded). She graduated in French and Comparative Literature at the Université Sorbonne and specialises in the relations between music and literature. She completed her PhD thesis in Comparative Literature, entitled ‘Bodies in Composition: Women, Music, and the Body in Nineteenth-Century European Literature’ at the University of Kent and at the Université Sorbonne in 2016. As part of the Baudelaire Song Project, she researches more particularly song settings by female composers as well as adaptations of Baudelaire’s poems in French rap music.

Tillman Weyde is a Senior Lecturer at the Department of Computing.  He currently works on Semantic Web representations for music, methods for automatic music analysis, audio-based similarity and recommendation and general applications of audio processing and machine learning in industry and science.

 
 
 
 
 
 
  • Black Twitter Icon

© 2019 University of Oxford