This article seeks to discuss metadata in the context of sound archives. It is informed by ongoing research at The I.S.R.O to build LEKHA.CC, an art archival platform from the ground up. Building on The I.S.R.O’s work in sonic research, special emphasis has been placed on sound, as a medium, to be significantly represented on LEKHA from the beginning. LEKHA thus aims to democratise archiving as a process through a usable and accessible platform, which can support a myriad of file formats.
This post was written by Nithya Kirti with the help of Arkoprabho Bhattacharjee, research assistants at The I.S.R.O
According to Merriam-Webster, 'Metadata' is simply ‘data that provides information about other data.' The term is widely used in the context of archiving to refer to the information that describes a particular artefact. For example, the fact that Mother Teresa was painted by Maqbool Fida Hussain (Creator) sometime around 1988-89 (Date Created) and is made with Oil on Canvas (Type) are all metadata details, which provide context to the actual painting (the artefact). This painting is furthermore represented online on museum websites as a scan/photograph, which is its digital surrogate.
In the context of sound and music, metadata is described as the information that is embedded in an audio file and used to identify content. Metadata, in this context, is crucial since it shows clear ownership over a track and credits the relevant artists/creators. Furthermore, it ensures proper documentation of a wide range of details about objects (descriptive, physical, structural), such as the bit rate, sampling and other technical information about files.
The first music catalogues from the late 1890s hardly ever listed the names of performers and composers; prospective buyers were more curious about the title of a song or an opera. Modern music and sound records feature a detailed discography, or "music metadata," as their content becomes richer with time and subsequent catalogues offer much more detail. All those details about a song that show up on music services or when you search for details of an audio file are 'metadata'.
This would include-
the track title,
track number, etc.
Furthermore, these catalogues offer the opportunity to explore a highly complex iconography- of photographs, drawings, etc. that is complementary to the sounds themselves rather than simply being a monotonous list of artists in alphabetical order.
Ethnomusicology is the interdisciplinary field of studying music in its social and cultural contexts. An ethnographic sound archive, thus, contains audio recordings and/or their digital surrogates, along with contextualising documents- field notes, photographs, field footage, etc. Sound archives, thus, understandably play an integral role in enabling research in ethnomusicology.
Established in 1899 by members of the Imperial Academy of Sciences, the Vienna Phonogrammarchiv was the first sound archive in history. Over its nearly 100-year history, the Phonogrammarchiv's collections have expanded to over 50.000 recorded items, or around 7000 hours of content.
Together with the historical collections of the Berlin Phonogrammarchiv, they make up one of the few early collections of phonographic field recordings with worldwide coverage that illustrate stages of orally transmitted civilizations before the impact of western civilisation, some of which have gone extinct since the time of recording or have at least undergone significant change., The British Library Sound Archive is currently one of the largest and most extensive sound archives in the world.
For the International Association of Sound and Audiovisual Archives (IASA), a "sound recording" includes:
audio formats (published, unpublished and broadcast);
moving image formats, where these are considered to be a natural extension of audio formats (e.g. music videos, musical performances on laserdisc) or related to audio (e.g. simultaneous FM radio and television broadcasts); and
electronic resources (e.g. interactive CD-ROMs, audio content in jukeboxes and mass storage systems)
A piece of music, thus, otherwise streaming on a music service platform, could thus become archival material if it is catalogued in a way that gives equal importance to its metadata (descriptive, structural and administrative) and contextualises its creation and significance.
In sound art and audiovisual records, the album to file to item relationships are significant and need to be properly accounted for.
With each passing era, the cataloguing of sound has become more complex due to changing technologies of sound recording. There is an understandable move towards digitizing analogue recordings, such as wax cylinders from yesteryears, to aid in preservation as well as make content accessible to a larger audience. While the faithfulness of any digitization to the original is a question of technology, the end goal is to at least have detailed and accurate metadata for artefacts, that may fill in possible gaps.
Furthermore, newer challenges come up when considering digital archiving, especially of sound files that are 'born digital' i.e. files that were created digitally. Considerations of copyright and licenses become essential in scenarios where the digital files being archived are the original archival material.
The question now is- how does one go about accurately archiving sound artefacts?
Sound Metadata Standards
Like any other archival material, audio artefacts require careful cataloguing based on standardised metadata fields and controlled vocabularies, to find consistent anchoring in archives. The question (or rather, the challenge) is to decide whether to have specialised standards for sound or adapt and extend existing standards to accurately record the complexities that come with sound. Some options for existing standards for sound are as follows.
The International Association of Sound and Audiovisual Archives (IASA) Cataloging Rules specify requirements for the description and identification of sound recordings and related audiovisual media, assign an order to the elements of the description and specify a system of punctuation for that description. They are designed for use by sound and audiovisual archives (i.e. specialist audiovisual archives) as a guide in the preparation of cataloguing records and as a standard for the exchange of bibliographic information concerning sound and related audiovisual materials.
The overarching fields included under this are-
1. Title and statement of responsibility
2. Edition, issue, etc.
3. Publication, production, distribution, broadcast, etc., and date(s) of creation
5. Physical description
8. Numbers and terms of availability
Read more here.
The Metadata Encoding and Transmission Standard (METS) schema is a standard for encoding descriptive, administrative, and structural metadata regarding objects within a digital library, expressed using the XML schema language of the World Wide Web Consortium.
METS provides a standardised framework in the form of an XML wrapper for durably linking the different types of metadata with the digital object. The digital object, for example, may have several physical formats-a high-resolution digital preservation master/original and a low-resolution web display version/digital surrogate. METS links all the metadata and all the physical instances of a single object together into an XML package that can be transported between METS-compliant systems. Read more.
METS thus allows for great interoperability between various standards. It is particularly helpful in expressing relations among physical and digital objects encompassing an artefact.
audioMD and videoMD are XML Schemas that detail technical metadata for audio- and video-based digital objects. They often serve as extension schemas within the METS administrative metadata section. Read more.
They are also suitable for use as standalone metadata documents or may be considered for incorporation into other structures, e.g., as embedded metadata in Material eXchange Format (MXF) files.
For Non-Specialist Archives
Adopted from ‘Best Practice Guide to Film and Sound Archives in Non-Specialist Repositories’, published by The Society of Archivists, the following fields represent basic metadata fields that can be adopted for all archival material, with specifications on how to adapt the fields for audiovisual media. Thus, the way sound art can be catalogued is thus dependent on the controlled vocabularies and the information that is inputted into the fields.
Reference Code (usually the accession number with sub-number)
Title (only that within the item itself, otherwise create one)
Date(s) (when item made or transmitted; may need to be
Physical Form (whether a cinefilm, video or sound recording)
Technical Description (carrier, gauge, size, base material, etc.)
Extent (Number of components, or part of a compilation)
Location (shelf in strongroom)
Copyright & Access (reference to details of rights owner/s, licensing
Provenance (source of item and proprietary ownership;
Documentation (written material accompanying item)
Content Analysis (synopsis, preferably with timings)
The following very basic recommendations by IASA are intended as a first step, a collection of data which is necessary to manage the file, or which must be captured or it would otherwise be lost:
Unique Identifier: Should be structured, meaningful and human-readable as well as unique. A meaningful identifier can also be used to relate objects, like master or preservation files and distribution copies, metadata records, series, etc where a sophisticated system will manage that in the metadata.
Description: Description of the sound sequence. A small amount of text to simply identify the content of the audio file.
Technical Data: Format, sampling rate, bit rate, file size. Though this information can be acquired later, making it an explicit part of the record allows for management and preservation planning of the collection.
Coding History: In BWF many discrete lines of information describing the original item and the process and technology of creating the digital file that is being archived. (See also 3.1.4 Metadata).
Process errors: Any error data which the transfer system can collect which describes failings in the transfer process (e.g. uncorrectable errors in CD transfers).
Thus, to conclude, the landscape of sound archiving is as dynamic as it gets. Besides preservation challenges, sound artefacts require minute details, especially regarding the technical and contextual aspects of the audio and links with other artefacts, to be recorded in order to present the full picture (or sound, in this case) and reflect their unique, situated contexts.
Consequently, the metadata to be adopted to enable such accuracy depends on the institutional/individual requirements.
The usual challenges to accessibility that come with sound recordings, such as language barriers, inaccuracies in translation, lack of transcripts, background information regarding the subject of audio, etc. remain and are as important as ever. An ideal metadata plan should allow for such challenges to be structurally addressed as well.