Skip to main content

About this book

The future of music archiving and search engines lies in deep learning and big data. Music information retrieval algorithms automatically analyze musical features like timbre, melody, rhythm or musical form, and artificial intelligence then sorts and relates these features. At the first International Symposium on Computational Ethnomusicological Archiving held on November 9 to 11, 2017 at the Institute of Systematic Musicology in Hamburg, Germany, a new Computational Phonogram Archiving standard was discussed as an interdisciplinary approach. Ethnomusicologists, music and computer scientists, systematic musicologists as well as music archivists, composers and musicians presented tools, methods and platforms and shared fieldwork and archiving experiences in the fields of musical acoustics, informatics, music theory as well as on music storage, reproduction and metadata. The Computational Phonogram Archiving standard is also in high demand in the music market as a search engine for music consumers. This book offers a comprehensive overview of the field written by leading researchers around the globe.

Table of Contents




Computational Music Archiving as Physical Culture Theory

The framework of the Computational Music and Sound Archive (COMSAR) is discussed. The aim is to analyze and sort musical pieces of music from all over the world with computational tools. Its analysis is based on Music Information Retrieval (MIR) tools, the sorting algorithms used are Hidden-Markov models and self-organazing Kohonen maps (SOM). Different kinds of systematizations like taxonomies, self-organazing systems as well as bottom-up methods with physiological motivation are discussed, next to the basic signal-processing algorithms. Further implementations include musical instrument geometries with their radiation characteristics as measured by microphone arrays, as well as the vibrational reconstruction of these instruments using physical modeling. Practically the aim is a search engine for music which is based on musical parameters like pitch, rhythm, tonality, form or timbre using methods close to neuronal and physiological mechanisms. Still the concept also suggests a culture theory based on physical mechanisms and parameters, and therefore omits speculation and theoretical overload.
Rolf Bader

Fieldworks and Archives


“The Lanang Is the Bus Driver”: Intersections of Ethnography and Music Analysis in a Study of Balinese Arja Drumming

This chapter presents an ethnographically informed analysis of the improvised Balinese drumming practice kendang arja, blending a personal and sometimes informal ethnographic writing style with close musical analysis. In this paired drumming tradition, two musicians, without formal music theory to guide them, somehow create the tightly interlocking rhythms so characteristic of Balinese music through continuous and simultaneous improvisation. No surprise these are among Bali’s most respected musicians. By examining the improvisations of various master drummers via the lens of their informal oral music theory, this chapter seeks to make explicit several relatively implicit guidelines and techniques for kendang arja improvising. It thus offers insight into the ways these musicians create mutually compatible patterns in the course of performance, while simultaneously demonstrating the value of ethnographic fieldwork to the analysis of improvised musics.
Leslie Tilley

Temperament in Tuning Systems of Southeast Asia and Ancient India

Tuning systems in many musical cultures in Southeast Asia as well as in India are often considerably different from Western tuning systems. Within these cultures, the literature as well as oral tradition on theoretical reflections of why choosing which tuning system are scarce. For Western scholars, only considering the pitches played or the tuning of instruments is not perfectly satisfying, as it lacks insight into the reasons for certain tunings. Still when considering ecological constraints, like ensemble setups, instrument building, or the acoustics involved, often the reasons for choosing a special tuning becomes clear. The paper presents examples from fieldwork in this regions between 1999 and 2014, and often finds musicians wanting to get close to just intonation, but deviating from it mainly because of three constraints: The need to switch between different ensemble types, constraints of instrument building, and acoustical constraints. The paper therefore suggests that many tunings in Southeast Asia are temperaments in the sense of compromises between a desired system, often just intonation, and such constraints.
Rolf Bader

John Blacking Revisited—Comparative Analysis of Venda Tshikone Dance (1958 and 2009)

One of the first research topics of interest for folk music researchers was the stability/instability of folk tunes. Particularly the question which parts and elements in folk tunes are sensitive to change and what are the stable elements of the melodies. Researchers had data collected over several decades, which gave a great opportunity to systematically explore stability of melodies over the years and decades.
Jukka Louhivuori

Smithsonian Folkways and the Associated Ralph Rinzler Folklife Archives and Collections

Smithsonian Folkways and the associated Ralph Rinzler Folklife Archives and Collections were created in 1988 as a vehicle to collect world music record labels and disseminate the recordings in a number of ways. In addition to the recordings of the 50 years of the Smithsonian Folklife Festival which has included participants from all over the world, the Smithsonian Institution has been collecting the recordings of independent record companies, 13 to date. This includes Folkways, Arhoolie, Paredon, Cook, the Mickey Hart World Music Collection and the UNESCO labels. The plan of collecting record labels allows us to gain all the intellectual property rights. This assures we can legally distribute the recordings and see that the musicians, informants, and communities receive their due royalties. The recordings as distributed as commercial CDs or LPs, streaming audio, or downloads. One can download tracks through our website or services like Itunes. We have created over 400 new recordings since 1988 and have over 4000 recordings in our catalog. Our robotic Micro-Tech machine can create compact discs of the 4000 titles. The roughly 4400 commercial recordings are about 15% of what exists in the Rinzler Archives. The rest consists of festival recordings and outtakes from the recording labels. This paper hopes to share a brief glimpse into our processes.
Jeff Place

Analysis and Perception of Javanese Gamelan Tunings

Gamelan music as performed on the Indonesian islands of Java and Bali is one of the most well known and well studied non-European music traditions. Especially the tunings of gamelan ensembles have fascinated researchers since the early stages of the disciplines of Systematic and Comparative Musicology, as these tunings are considerably different from Western, Middle Eastern or other music traditions and even differ between gamelan ensembles. Additionally most instruments of the ensemble are percussion instruments with inharmonic overtone spectra. One way to explain certain characteristics of gamelan tunings is by relating the inharmonic sound spectra of the percussion instruments to the perception of dissonance. This theory is assuming that the psychoacoustic sensation described as auditory roughness is perceived as dissonant and is therefore avoided. This study investigates the influence of musical roughness on the perception of different gamelan tunings by correlating psychoacoustic measurements with a perception test in an online survey. Based on sound samples of an existing gamelan ensemble set based in Hamburg, Germany, a gamelan tune was built in a DAW. By detuning the sounds, several versions of the tune in different temperaments were built. It appears that the measured roughness perception and roughness measurements of the different tunes correlate very well for all detuned cases. Still the original piece does not correlate, pointing to a different perception strategy for the original ensemble tuning.
Gerrit Wendt, Rolf Bader

Computation, Networking and Platforms


Content-Based Music Retrieval and Visualization System for Ethnomusicological Music Archives

In this chapter we propose a content-based exploration and visualization system for ethnomusicological archives that allows for data access by rhythm similarity. The system extracts an onsets-synchronous timbre feature of each audio file of a given collection. From the resulting time series, Hidden Markov Model are trained. The transition probability matrices of the models are considered a rhythm fingerprint that represents the musics rhythmic structure in terms of timbre. The self-organizing map algorithm is utilized to project the high-dimensional fingerprints onto a two-dimensional map. This technique preserves the topology of the high-dimensional feature space, which results in similar map positions for similar rhythms. A clustering by rhythm similarity is thus achieved. The system, therefore, supports musicologist studies in several ways: the rhythm fingerprinting does neither imply a certain theory of music nor introduce cultural bias. Hence, different musics can be compared meaningfully regardless of their origin. Retrieval by similarity allows for an explorative approach to the music collections, which can support researchers in finding new hypothesis and utilizing music archives with few or without meta data. The system is currently prototyped in the Ethnographic Sound Recordings Archive of the University of Hamburg as a part of the COMSAR project.
Michael Blaß, Rolf Bader

Spatial Manipulation of Musical Sound: Informed Source Separation and Respatialization

“Active listening” enables the listener to interact with the sound while it is played, like composers of electroacoustic music. The main manipulation of the musical scene is (re)spatialization: moving sound sources in space. This is equivalent to source separation. Indeed, moving all the sources of the scene but one away from the listener separates that source. And moving separate sources then rendering from them the corresponding scene (spatial image) is easy. Allowing this spatial interaction/source separation from fixed musical pieces with a sufficient quality is a (too) challenging task for classic approaches, since it requires an analysis of the scene with inevitable (and often unacceptable) estimation errors. Thus we introduced the informed approach, which consists in inaudibly embedding some additional information. This information, which is coded with a minimal rate, aims at increasing the precision of the analysis/separation. Thus, the informed approach relies on both estimation and information theories. During the DReaM project, several informed source separation (ISS) methods were proposed. Among the best methods is the one based on spatial filtering (beamforming), with the spectral envelopes of the sources (perceptively coded) as additional information. More precisely, the proposed method is realized in an encoder-decoder framework. At the encoder, the spectral envelopes of the (known) original sources are extracted, their frequency resolution is adapted to the critical bands, and their magnitude is logarithmically quantized. These envelopes are then passed on to the decoder with the stereo mixture. At the decoder, the mixture signal is decomposed by time-frequency selective spatial filtering guided by a source activity index, derived from the spectral envelope values. The real-time manipulation of the sound sources is then possible, from musical pieces initially fixed (possibly on some media like CDs), and with an unpreceded (controllable) quality.
Sylvain Marchand

Finding Music in Music Data: A Summary of the DaCaRyH Project

The international research project, “Data science for the study of calypso-rhythm through history” (DaCaRyH), involved a collaboration between ethnomusicologists, computer scientists, and a composer. The primary aim of DaCaRyH was to explore how ethnomusicology could inform data science, and vice versa. Its secondary aim focused on creative applications of the results. This article summarises the results of the project, and more broadly discusses the benefits and challenges in such interdisciplinary research. It concludes with suggestions for reducing the barriers to similar work.
Oded Ben-Tal, Bob L. Sturm, Elio Quinton, Josephine Simonnot, Aurelie Helmlinger

Experimental Investigations and Future Possibilities in Network-Mediated Folk Music Performance

This chapter is intended to acquaint music ethnologists with the paradigm of Networked Music Performance (NMP). NMP facilitates computer networks to allow musicians from distant geographic locations to synchronously collaborate during performance, improvisation or more generally music-making. The chapter comprises two parts. The first part is devoted to providing an overview of research approaches in NMP and elaborates on the technical and perceptual impediments restricting the wide availability of this type of technology. The second part presents an experiment involving three musicians performing folk music over the network. The experiment serves to reveal not only technical and perceptual difficulties in the communication of performers, but more importantly their attitude towards engaging in this novel practice. The chapter concludes by discussing future perspectives on the use of NMP technology in the context of ethnic and folk music.
Chrisoula Alexandraki

Requirements and Use Cases for Digital Sound Archives in Ethnomusicology

In this chapter, results from requirements elicitation activities for an ethnomusicological sound archiving software are summarized. Results derived from user and expert interviews as well as from literature of the sphere of ethnomusicology, computational musicology, archival studies and informatics. Interviews were conducted with stakeholders that are either involved in creation and maintenance of archiving software or who consider utilizing an archive software for private or professional use. Particular emphasis was placed on requirements and use cases that are supported by recent digital technologies. Online publishing, distributed systems and computational analysis of audio content and metadata can enable ethnomusicological sound archives to bring new value to their corpora. Publishing and sharing contents online can extend academic and private uses and computational analysis of archive contents can help to structure, access and maintain archives in novel ways. As a final result of requirements elicitation activities, technology, architecture and features of an archiving software built based off the findings are briefly presented and future tasks and challenges for this specific implementation are outlined.
Jonas Franke

Physical Modeling and Measurements


Laser-Based Interferometric Techniques for the Study of Musical Instruments

Laser-based interferometry techniques for the study of musical instruments are discussed in this chapter. The presented work demonstrates the advantages that the laser-based optical techniques provide and is mainly focused on the capabilities of the Electronic Speckle Pattern Interferometry (ESPI). The mathematical description of the time-average ESPI, the reading of the contour lines and the overcoming of the limitations of amplitude and phase vibration are analyzed. Furthermore, four representative studies using the ESPI for a Cretan Lyra, a Bendir, a classical Guitar and the ancient Greek lyra Chelys, are presented and demonstrate the capabilities of the method.
Efthimios Bakarezos, Yannis Orphanos, Evaggelos Kaselouris, Vasilios Dimitriou, Michael Tatarakis, Nektarios A. Papadogiannis

Shock Wave Characteristics in the Initial Transient of an Organ Pipe

A new approach to investigate the role of pressure waves in the initial transient of an organ pipe is presented. By numerical simulations solving the compressible Navier-Stokes equations with suitable boundary and initial conditions, it is possible to retrace the generation, propagation, reflection, damping and radiation of sound waves. The focus is on the contribution of occurring pressure wave fronts in the initial transient that show shock wave characteristics, in particular their role at the formation process of the sound field inside the organ pipe’s resonator. Utilizing spectral analysis as well as extended visualization methods, a wide range of aspects of the dynamics of the initial transient of an organ pipe is discovered. In particular the damping processes in the resonator which are nonlinear are analyzed and discussed in detail. The numerical approach presented in this case study, allows to study the initial transient of an organ pipe with an extraordinary level of precision, thereby helping to understand the underlying first principles of the sound field formation and the mutual interaction of the jet’s flow field and the sound field inside organ pipes and similar wind instruments that produce complex sounds listeners find both interesting and joyful. Animations of the temporal and spatial development of relevant physical quantities like pressure, turbulent kinetic energy, vorticity and velocity magnitude calculated in the numerical simulations are provided as supplementary material.
Jost Leonhardt Fischer

Computed Tomography as a Tool for Archiving Ethnomusicological Objects

Musical instruments in ethnological collections can be a challenge for museums. Objects with uncertain provenance or doubtful circumstances of acquisition are considered to be repatriated. Some objects consist of sensitive material like human remains and are therefore bound to ethical guidelines for exhibition. On the example of a Tibetan damaru, a drum made of two human skulls, the provenance of the object and ethical considerations are discussed. For the case of repatriation 3D computed tomography is presented as a powerful examination and archiving method. Furthermore, virtual presentation and research concepts as well as other museum applications of 3D data of musical instruments are considered.
Sebastian Kirsch

3D Imaging of Musical Instruments: Methods and Applications

This treatise is intended as an introduction to three-dimensional (3D) imaging for stakeholders working with musical instruments, e.g. ethnologists, musicologists, curators, and instrument builders. The work should help to find the appropriate method for a specific purpose and further highlight several possible applications for the obtained data. Firstly, three techniques of 3D image acquisition are introduced. Advantages and disadvantages of the proposed methods are discussed in terms of ease of use and obtained information. Secondly, a workflow is presented to post-process the captured raw data. Finally, several examples of possible utilization of the generated virtual models are introduced. As far as possible, the proposed procedures are based on the use of open source software/freeware and should be applicable on regular current personal computers. Parts of this work are based on a proceedings abstract published in 2017 [29].
Niko Plath

How to Interprete Early Recordings? Artefacts and Resonances in Recording and Reproduction of Singing Voices

Voice recordings in the beginning of the 20th century required a complex acoustic and mechanical set-up for transfer of the singing voice sound from singer via horns, ducts, soundbox and needle to a cylinder or disc. A similar construction was used to reproduce the voice. Both signal paths had significant impact on various properties of the voice signals—mostly the voice signal quality was reduced and distortions were produced. Another effect were changes in details of the voice spectrum due to the interaction of resonances in the voice signal and the transfer paths of recording and reproduction devices. We present analyses of the voice signal modifications and relate them to parts of the devices. This work is funded by the German Research Foundation.
Malte Kob, Tobias A. Weege
Additional information

Premium Partner

image credits