The following is a reflection that was created for GLIS 689 Selected Topics. This course was offered by the School of Information Studies at McGill University in 2019.
The Presbyterian Church in Canada Archives has a collection of recorded interviews on audio cassettes that date from 1977 to 2010. The Church’s Committee on History has encouraged church congregations to conduct interviews with ministers, missionaries, and church workers to understand their experiences working for the Church. These interviews also serve as an oral history of synods and ministries that comprise of The Presbyterian Church in Canada. The Presbyterian Church in Canada Archives would like to process and preserve these records and make a selection of the recordings available on their website.
To begin this project, I worked with my supervisor Robert Anger to develop a workflow for the digitization process. This workflow included re-housing and storage, arrangement and indexing, and identification and description. I also constructed an annotated bibliography that explored current theories, issues, and practices related to oral history archiving. As part of this project, I toured the Archives of Ontario to view their audio-visual collection and understand their best practices. The Archives of Ontario digitizes audio cassettes in a similar way to what I proposed in this project. Currently, the Archives of Ontario use VEGAS Pro for capturing audio and Adobe Premiere Pro to edit digital audio files. The Archives of Ontario ingests audio at a sampling rate of 48kHz. Previously, they used Audacity as a capture and editing software. They opted for VEGAS Pro and Adobe Premiere Pro for its advanced capturing and editing capabilities.
I continued the project by adhering to the workflow. The collection included 31 audio cassettes, 3 microcassettes, 6 video cassettes, 5 CDs, 2 DVDs, 1 minidisk, and 15 documents. I created an inventory of the audio cassettes in the collection and placed the cassettes into archival boxes. The audio cassettes were inspected to check for damage and rehoused in new casings. The other recordings were separated from the collection to be catalogued separately based on the type of audio-visual material. The write protection tabs on the audio cassettes were removed to make sure the tapes were not accidentally recorded over. An external audio cassette player was used to play the audio cassettes and Audacity was used to capture and edit the recordings. To begin the digitization process, the equipment was tested and configured using a demo cassette. The audio was ingested at a sampling rate of 48kHz with and a bit depth of 24-bit. Once digitized, the audio tapes were saved as uncompressed Waveform Audio Format (WAV) files for preservation copies and compressed Moving Picture Experts Group Layer-3 Audio (MP3) files for access copies. Audacity was also used to embed metadata in the audio recordings. “Artist name” was used to record the name of the interviewer, “track title” was used to record the name of the interviewee, and “year” was used to record the date of the interview.
Items that had little or no identification were identified and described by listening to the recording in full. At the file level, the recordings were arranged by interviewee name. At the item level, the recordings were arranged numerically by the date of the interview. Information was entered into an Inmagic DB/TextWorks database that was newly designed for audio-visual recordings. The recordings were cataloged and described following the Rules for Archival Description (RAD) for sound recordings. Documentation that accompanied the recordings were scanned, saved as PDFs and catalogued in Inmagic DB/TextWorks with its relationship to the audio recordings clearly expressed. As none of the audio-visual materials in the archive were cataloged, a naming convention for these materials was created. The naming convention was created to specify the type of audio-visual material. For example, the audio cassettes were given unique identifiers following AV-1-cassette. Similarly, DVDs were given identifiers that followed AV-1-DVD, born digital files were given AV-1-digital, Reel-to-reel tapes were given AV-1-reel-to-reel, CDs were given AV-1-CD, and VHS tapes were given AV-1-VHS. Digital files of the audio cassette recordings were named following its unique identifier and the name of the interviewer, the name of the interviewee, and the date created. For example: AV-1-cassette Bob Anger Interview with James Smith 8 November 2000.wav. This naming convention indicates the contents of the file, the date of creation, and the structural relationship between files. Similarly, documents that accompanied the recordings were named following the unique identifier and a description of the type of document. For example, interview transcripts for an interview for James Smith conducted by Bob Anger would be named AV-1-cassette Interview transcripts.pdf. Digital files were store on an external hard drive. The documentation was physically arranged separately from the audio recordings.
Limitations and challenges
There were a few limitations to this project. As previously mentioned, the collection included 3 microcassettes, 6 video cassettes, 5 CDs, 2 DVDs, and 1 minidisk. These materials were separated from the collection to be catalogued separately based on the type of audio-visual material. As the archives was only equipped with an external audio cassette player, interviews on microcassettes, video cassettes, and minidisk could not be digitized. To begin the project, the Oral History Metadata Synchronizer (OHMS) program was explored as a free-open source option for cataloging and storing oral history interviews. OHMS is useful for supporting metadata, keyword search, access, transcriptions, and maps. For the scope of this project however, I did not have the time to create transcriptions or develop an account on OHMS. Inmagic DB/TextWorks was chosen as an alternative to OHMS to catalog the recordings, as it was already being used by the archives, and an external hard drive was used to store the recordings. Given the time limitations, the recordings were not made accessible on The Presbyterian Church in Canada Archives website. The Presbyterian Church in Canada Archives intends to make a selection of the recordings accessible on their website in the future.
For anyone who is interested in creating their own oral history or audio archive, there are a few recommendations for best practices. You should place cassettes in new cases if the original housing is missing, acidic or broken, otherwise you can leave the tapes in their original housing. Remove the write protection tab to make sure the tapes are not accidentally recorded over. Be sure to record any information with names, dates, places, or any other contextual information into the new cases. It is recommended to use Polypropylene audio cassette cases. The audio cassettes should be stored vertically in clearly labeled audio cassette boxes. To improve longevity, the tapes should be stored at temperatures between 8 and 23˚C (Hess, Iraci, & Flak, 2012). The audio cassettes should be inspected to check for mold or damage. Playing damaged or moldy tapes can cause damage to the tape or the playback equipment. Professional services can be used to restore the tapes.
You should consider capturing high quality audio recordings as a preservation action. If possible, audio should be recorded at a minimum sampling rate of 48kHz with 96kHz preferred, and a minimum bit depth of 24-bit with 32-bit preferred. Otherwise, it is acceptable to ingest audio at a sampling rate of 44.1kHz and bit depth of 16-bit. Once digitized, audio tapes should be saved as uncompressed Waveform Audio Format (WAV) or Broadcast Wave Format (BWF) files for preservation copies. It is acceptable to save audio files as compressed Moving Picture Experts Group Layer-3 Audio (MP3) files for access copies (Hess, Iraci, & Flak, 2012). Audio can be digitized by using an external audio cassette player that can be connected to a computer by USB, in combination with an audio capture software such as Audacity to capture and edit files. Digitizing audio can be time consuming as it will take the length of the tape plus edit and export time. Capture a few moments of silence in beginning and end of the recording and edit it out later so that portions of the recording are not missed. Meaningful file names should be given to the recordings to help with the management of the files. If you do not wish to, or are unable to digitize the audio yourself, there are public and private companies who provide digitization services. When researching companies, be sure that they follow best practices and have quality equipment for ingesting audio.
It is important to plan for obsolescence, upgrades, and file migration to ensure long term management and future access to your recordings. It is recommended to keep several back-ups of your recordings such as on a computer hard drive, an external hard drive, a CD, a DVD, and cloud services. When selecting a digital platform or cloud service, it is important to consider privacy and access issues. Digital platforms such as iTunes, YouTube, Oral History Metadata Synchronizer (OHMS) and Internet Archive’s Community Audio are useful and free services for storing audio recordings with descriptive information. Automated transcription software such as Google cloud speech-to-text, Trint, and Temi can be used to create transcriptions for the audio recordings. According to the National Digital Stewardship Alliance (NDSA) (n.d.), it is best practice to store at least 3 copies of your files and metadata in geographic locations that can be affected by different disaster threats. The original tapes should be retained in case of future digitization needs.
Hess, R., Iraci, J., & Flak, K. (2012). The Digitization of audio tapes – Technical Bulletin 30. Retrieved from https://www.canada.ca/en/conservation-institute/services/conservation-preservation-publications/technical-bulletins/digitization-audio-tapes.html
National Digital Stewardship Alliance. (n.d.). Levels of digital preservation. Retrieved September 10, 2019, from http://ndsa.org//activities/levels-of-digital-preservation/