2021 Proceedings

DLfM '21: 8th International Conference on Digital Libraries for Musicology

Full Citation in the ACM Digital Library

An archive-exploration system for the Hunting Songs of the Lakeland Fell Packs

  • Mary Emmett
  • Zongyu Yin
  • Tom Collins

From everyday users to expert archivists, there is a keen interest in being able to search digital library catalogues via domain-specific languages and queries. This paper presents a proof-of-concept system that enables library users to search musically for songs or pieces that match to sung, hummed, or typed note queries. We chose to work with a corpus of 80 Lakeland Hunting Song tunes (Items) that have been collected recently, and for which the archivist wanted to make a resource that was easily searchable in terms of its musical content. Existing systems such as that at folktunefinder.com are robust to rhythmic changes but not to missing or interpolated notes, so we employed a geometric hashing approach as the basis for determining matches and as a potential improvement on existing work. An evaluation indicates that the system works quickly and effectively. We discuss some of the examples arising from its use, and indicate how future work building on this proof-of-concept system may make it possible to search other collections of songs or pieces in a more natural, musical way. Our archive-exploration system is available for trying out at https://hse.glitch.me.

Archiving the Sonic Ephemeral: Towards a classification of sound installation documentations through spatial audio features

  • Nicolas D'Aleman Arango

This paper proposes the application of audio feature extraction for the classification of sonic documentations of the works presented during the Singuhr Hoergalerie Project from 1996 to 2014. More specifically, I utilize the Stereo Panning Spectrum Features to discern some of the relationships between the artworks based on the spatial characteristics of said documentations. The first part of this paper addresses the theoretical concerns that emerge from documenting a form of art that champions ephemerality as one of its core characteristics. Audio recordings force sound installations into an ontological reality and effectively positions them in the musical realm, thus subjecting them to a one-dimensional interpretation. An approach to overcome this paradox is built upon Kersten Glandien’s proposal to analyze sound art. The second part describes the technical processes of examining the Singuhr recordings through the proposed algorithm. The results present a novel form of approaching sonic documentations by analyzing spatial relations between audio sources, and how to incorporate them into additional documentary and archival material.

Cross-Cultural Corpus Creation and Statistical Tendencies in Music

  • Andrew Brinkman
  • David Huron

The notion that some musical features can be found in a majority of cultures across the globe has garnered scholarly attention within the past decade [6, 29]. However, the lack of both large and diverse musical corpora has made furthering work on this topic difficult. This paper addresses this issue by testing four purported statistical tendencies in melodic organization across four substantial samples of culturally diverse music. Three corpora include musical samples of European, Native American, and Chinese folk music. The fourth, a new corpus devised specifically for the purposes of this study, is a corpus of cross-cultural folk music including material from nearly 700 sampled audio recordings collected from 44 distinct cultures. The creation of this corpus involved establishing clear guidelines on how to parse audio materials, how to define a musical “phrase”, and how to transcribe musical information (e.g. pitch and duration).

While the primary purpose of this paper is to contribute to the dialogue on corpus creation techniques involving culturally diverse music, results for the testing of all four hypotheses are included. Results are consistent with a number of broad statistical tendencies in musical phrases that are evident above chance levels, and are common across the repertoires sampled. These include the tendency for small over large pitch movements, for large leaps to ascend, for musical phrases to fall in pitch, and for phrases to begin with an initial pitch rise. Limitations of our corpus creation methods and empirical tests are highlighted.

Notes on the Music: A social data infrastructure for music annotation

  • David M. Weigl
  • Werner Goebl
  • David J. Baker
  • Tim Crawford
  • Federico Zubani
  • Aggelos Gkiokas
  • Nicolas F. Gutierrez
  • Alastair Porter
  • Patricia Santos

Beside transmitting musical meaning from composer to reader, symbolic music notation affords the dynamic addition of layers of information by annotation. This allows music scores to serve as rudimentary communication frameworks. Music encodings bring these affordances into the digital realm; though annotations may be represented as digital pen-strokes upon a score image, they must be captured using machine-interpretable semantics to fully benefit from this transformation. This is challenging, as annotators’ requirements are heterogeneous, varying both across different types of user (e.g., musician, scholar) and within these groups, depending on the specific use-case. A hypothetical all-encompassing tool catering to every conceivable annotation type, even if it were possible to build, would vastly complicate user interaction. This additional complexity would significantly increase cognitive load and impair usability, particularly in dynamic real-time usage contexts, e.g., live annotation during music rehearsal or performance. To address this challenge, we present a social data infrastructure that facilitates the creation of use-case specific annotation toolkits. Its components include a selectable-score module that supports customisable click-and-drag selection of score elements (e.g., notes, measures, directives); the Web Annotations data model, extended to support the creation of custom, Web-addressable annotation types supporting the specification and (re-)use of annotation palettes; and the Music Encoding and Linked Data (MELD) Javascript client library, used to build interfaces that map annotation types to rendering and interaction handlers. We have extended MELD to support the Solid platform for social Linked Data, allowing annotations to be privately stored in user-controlled Personal Online Datastores (Pods), or selectively shared or published. To demonstrate the feasibility of our proposed approach, we present annotation interfaces employing the outlined infrastructure in three distinct use-cases: scholarly communication; music rehearsal; and rating during music listening.

Serial Analysis: A Digital Library of Rows in the Repertoire and their Properties, with Applications for Teaching and Research

  • Mark Gotham
  • Jason Yust

Recent years have seen the emergence of concerted efforts for consolidating and curating digital libraries of musical analyses. This is a welcome development that stands to complement the more established attention to collections of musical resources like scores, recordings, and manuscripts. This paper begins by presenting a new digital library of 12-tone rows used in the repertoire. The list incorporates and extends all previous scholarly and crowd-sourced efforts to provide human- and computer-readable information about the rows and their usage including composer, work, and date. We present this list in both textual and musical notation, and in both ‘fixed’ and editable formats to support the widest possible use and development of this resource across pedagogy and research.

Further, we present one initial application in each of those domains. First is a free, public-facing anthology as part of the Open Music Theory Textbook v.2, providing teachers and students with rows presented in terms of properties such as combinatoriality. Second is a substantial study of more novel row properties, using the Discrete Fourier transform to examine distributions of harmonic qualities in the rows, as well as trends by composer and date. Finally, to support replication and extension, we provide the list, anthology, and all the associated processing code used at https://github.com/MarkGotham/Serial_Analyser.

Towards a Foundation for Collaborative Digital Archiving with Local Concert-Giving Organisations

  • Charlotte Armstrong
  • Rachel Cowgill
  • Alan Dix
  • Christina Bashford
  • D. Stephen Downie
  • Mike Twidale
  • Maureen Reagan
  • Rupert Ridgewell

The centenaries of former chapters of the British Music Society (BMS), established in 1918, have prompted their governing bodies to take stock of their histories and build on the cataloguing, documentation and preservation of their archival collections. The InterMusE project aims to support this shared instinct to archive by capturing and, crucially, linking different forms of data regarding the musical events provided by three of these local concert-giving organisations, beginning with the digitisation of their collections and with a view to producing a dynamic, open-access digital archive. This paper outlines our approach to establishing a foundation for developing a new kind of digital archive for musicology that is both valuable for researchers, fulfils the needs of the societies and their communities, and sheds light on community music-making on a national and, ultimately, international scale. By carrying out a series of preliminary scoping exercises, including informal interviews and archival-collection assessments, we can compare current archiving and preservation activities across the societies. These conversations bring emerging themes, issues and challenges into focus, raising pertinent questions that will inform our development of transformative tools and techniques for community digitisation projects.

Narratives and exploration in a musicology app: Supporting scholarly argument with the Lohengrin TimeMachine

  • David Lewis
  • Kevin Page
  • Laurence Dreyfus

We present the Lohengrin TimeMachine web application, consisting of video and textual musicological essays supported by an interactive digital companion. The digital companion allows a user to browse and compare all the occurrences of a motive in the opera Lohengrin, viewing them by text, vocal score and orchestration, with detailed views, segment labelling, audio excerpts and textual commentaries supporting the exploration. The video and essay modes show live links into the companion as the viewer or reader progresses through the narrative. This application is built on Linked Data technology and demonstrates the viability of such an approach, with the knowledge graph being traversed in the user’s browser to gather the materials for display. It uses the Music Encoding and Linked Data (MELD) framework, which provides the basis for a range of music-related Linked Data applications.

In this paper, we describe and illustrate the application in use, its technological underpinnings, as well as the motivation and implementation experience.

The Path to the New Note: Interval Distributions in the Music of Anton Webern

  • Joshua Ballance

Intervals, the distances between notes, are a common topic in the analysis of post-tonal music, and this is certainly true with regard to the music of Anton Webern. There is a lack, however, of rigorous empirical analysis of this aspect of Webern’s music, especially on a large scale. Scholarship has instead tended to rest on unfounded assertions or localised phenomena, which cannot necessarily be generalised. This paper is a corpus study of Webern’s 31 works with Opus numbers, using music21 to analyse the interval distributions across the corpus and interrogate commonly-held assumptions about Webern’s music. In particular, this study traces changes in his intervallic language across his body of work, assesses the relationship between rows and the resulting dodecaphonic music, and considers the integration of his harmonic language from the perspective of correlations between vertical and linear intervallic content.

Symbolic Encoding of Simultaneities: Re-designing the General Chord Type Representation

  • Konstantinos Giannos
  • Emilios Cambouropoulos

Encoding note simultaneities (chords) has been approached in different ways, such as Roman numerals for tonal harmony, or pitch class sets encountered in atonal and non-tonal music. A novel chord representation, the General Chord Type (GCT) representation, was developed to be adaptable to a broad variety of harmonic idioms from tonal to atonal. Given a binary classification of intervals into consonant or dissonant, GCT rearranges the notes of a given simultaneity such that the base of the chord encoding is consonant. This study proposes a refined elegant version of the GCT algorithm that takes into account a graded raking of intervallic consonance, maintains the core characteristics of GCT and resolves known problems such as certain types of chord ambiguities (more than one chord types for a single pitch collection) and orderings of intervals (wrong base/root and chord extensions).

To evaluate the performance of the new version, common tonal chords, whose encodings are well established, are compared with encodings from both versions, as well with other existing encoding systems. In the novel algorithm, ambiguous outputs are reduced significantly, inversions are correctly identified, and foreign chord notes are organised towards the most dissonant top end of the encodings. The algorithm performs well in atonal contexts by encoding Tn-Types with great accuracy.

Encoding, Analysing and Modeling I-Folk: A New Database of Iberian Folk Music

  • Nádia Carvalho
  • Sara Gonzalez-Gutierrez
  • Javier Merchan Sanchez-Jara
  • Gilberto Bernardes
  • Maria Navarro-Cáceres

Folk music is a fundamental immaterial heritage that promotes cultural identity. However, it lacks a substantial body of open access materials, and its promotion has been disconnected from the education curriculum. In this context, facilitated access to annotated high-quality folk music content can promote better educational tools and enhance cultural heritage literacy. Based on this, we advance and detail three main contributions: 1) a standardized model to musically annotate Iberian folk music; 2) a new database, named I-Folk, with annotated files following the proposed model; and 3) tools for navigating and retrieving folk music contents from the database. A particular emphasis is given to the educational application of the proposed model, contents, and tools in education. Ultimately, we strive for the promotion of Iberian folk music to the educators’ community.

What are the most used guitar positions?

  • Jules Cournut
  • Louis Bigo
  • Mathieu Giraud
  • Nicolas Martin
  • David Régnier

The playing of stringed instruments can be notated with symbolic tablatures, indicating for each note, the string, and the position on the fretboard at which it has to be played. We propose an encoding for tablatures in which the hand position on the fretboard is encoded to ease computational analysis and generation of guitar music. We analyze a corpus of 1022 guitar tablatures in standard tuning that were accurately transcribed in the corpus MySongBook, spanning metal, rock, pop, jazz/blues, and baroque/classical/romantic repertoires. We show how some of the proposed encodings reveal relevant statistics and some invariance properties on such guitar music. We show study unusual guitarist gestures around the G/B string pair due to its interval of major third, compared to the perfect fourth between other adjacent pairs of strings. We study the corpus across genres and textures (rhythm and lead parts) and how same relative chord positions are used along the fretboard. We distribute sets of 1000 most frequent vectors in different encodings for these five music genres.

Discovering Common Practice: Using Graph Theory to Compare Harmonic Sequences in Musical Audio Collections

  • Jeff Miller
  • Vincenzo Nicosia
  • Mark Sandler

In recent decades, rapid technological advances have resulted in a huge quantity of readily accessible digital musical recordings. The scope of large corpora presents difficulties for curators but offers new opportunities to musicologists and music theorists. We propose an application of graph theory which enables comparison of harmonic content from musical audio across collections of recordings. We introduce a graph schema wherein the chord sequences of musical recordings are used to create directed, weighted graphs which represent the underlying harmonic structure of the source material. We believe this application of graph theory offers novel advantages over existing approaches: 1) the relative positions of the chords in the time domain are retained, allowing the graphs to represent entire harmonic sequences of musical material, and 2) sequences from multiple sources are combined into a single graph, exposing features which are common to the source musical material, but which may vary or be absent from any particular instance. To test the schema, graphs were generated from recordings of ‘Georgia on My Mind’. We were able to produce examples demonstrating how this schema could be used to identify the essential harmonic framework of the song, to gain insight regarding the usage of chord substitutions by an artist during a single performance of the song, and to compare chord choices by two artists representing two different genres.