In my previous post, I mentioned that audio-video recordings are one of the most underutilized forms of historical sources according to Michael Frisch. In the coming years after Frisch wrote Oral History and the Digital Revolution, this would remain true. In 2010, the Council on Library and Information Resources and the Library of Congress released a report (The State of Recorded Sound Preservation in the United States: A National Legacy at Risk in the Digital Age) that claimed cultural heritage institutions would stop preserving sound archives if students and scholar did not start utilizing them. In response, the School of Information at the University of Texas at Austin and the Illinois Informatics Institute at the University of Illinois at Urbana-Champaign decided to create the High Performance Sound Technologies for Analysis and Scholarship (HiPSTAS). Originally funded by the NEH, archivists, librarians, scholars and students came together to discuss how audio collections could be made more accessible and how to create a suite of open-source tools to foster such scholarship. Through this dialogue, all participants from affected fields were able to relay their concerns with organizing as well as accessing sound artifacts.
According to HiPSTAS’s grant proposal, “there [was] no provision for scholars interested in spoken texts such as speeches, stories, and poetry to use or to understand how to use high performance technologies for analyzing sound” (1). This can be seen in Wendy F Hsu’s Digital Ethnography Toward Augmented Empiricism: A New Methodological Framework, as she hoped to, “spark[ed] some interest in creative engagement with digital methods in ethnography,” by explaining a methodological framework for using audio files. Writing in 2014, Hsu saw the potential for work with audio files and felt similarly as the people behind HiPSTAS that some sort of structure should exist for first time users of these audio sources. In particular, Hsu focuses on sound-based cultures that can be found on the internet. She listed a lot of different sites and resources where she conducts her research, which was interesting since Hsu mentioned our two practicums for this week, Audacity and SoundCloud. As Hsu states, “Digital technologies afford us the capability to engage with empirical data in multiple modes and at multiple levels.” By examining bites of sounds through these technologies, there is potential for deeper cultural understandings. However, this can only be accomplished if there is an understanding across the board by researchers and cultural heritage institutions of methods of best practice to utilize audio collections. This was the main purpose behind HiPSTAS.
Since 2013/2014, HiPSTAS has come through on its promise to offer guides and tools to work with audio. Introducing the HiPSTAS Audio Toolkit Workflow: Audio Labeling by Tanya Clement reviews the new application by HiPSTAS mentioned in the title and the workflow for deconstructing complicated audio files. By creating tools for students and scholars to use, the task of searching through audio files will not seem so daunting as long as the steps that Clement provides are followed. In this particular application, the audio project focused on distinguishing sounds, from the main voices to background noises such as applause. By inventing technologies to process the distinctions between these sounds, oral histories will gain a clearer focus in instances where numerous sounds exist on one track. As this project was from just a year ago, it appears that HiPSTAS met its goals and cultural heritage institutions will preserve sound archives as students and scholars start utilizing oral histories more in their research.
Questions
- Have you attended any conferences or taken any classes where oral histories were used? How were the used, explained, or described to you?
- Do you think HiPSTAS met their goals? Is there still more work to be done?
- When reading Clement’s steps, did the workflow make sense to you? Do you think you could conduct research on sound collections with this knowledge?
This HiPSTAS website was particularly interesting to me because I have been dealing with audio files a lot this semester. I am doing a project for another class where we are working with audio files from NPR. NPR recently received a grant (although I can’t remember exactly where they received it from) which allowed them to digitize many of their episodes of “All Things Considered” from the 60’s and 70’s. They really want to make their audio publicly available but they are facing a lot of copyright issues surrounding the audio. Because some reports come in from the BBC or freelance reporters they often don’t have rights to some of the material. This is interesting to me in terms of accessibility because even if places are able to digitize audio and take advantage of HiPSTAS grants , there are often still obstacles to distribution. This is not to mention the fact that scholars often will not even turn to audio as a form of historical analysis.
After learning about oral history for the first time in a class dedicated to the subject I couldn’t believe how horribly mistreated oral histories are in other classrooms. I’ve been given straight transcriptions (without any context or descriptors of the interviewer or interviewee!). I’ve had a professor talk down about the validity of oral history, and have dismissed the field as anything other than storytelling. I am not surprised by any of this, and students were or are not utilizing oral histories because they aren’t taught how to understand them. Lina’s project she talked about above with NPR is a huge step for the field – by validating audio as usable primary spruce material it sets us on a clear path into helping others understand that narratives are important, not matter the format they come in.
Oral history is interesting to me because it doesn’t actually apply to my own research at all. I generally focus on pre-1900 topics, so other than the WPA Slave Narratives (which have never directly related to anything on which I’ve worked), not a lot of oral history is relevant to my projects. So I’ve never taken a course on it, and it’s really interesting how much of a separate thing it is. I was even surprised to find out from a friend that oral histories are (for some reason I haven’t fully looked into) exempt from usual IRB requirements.
I think that one of the major challenges in using sound-based primary sources is just the searching process. Unless there is a transcript it’s much harder to search beyond broad keywords. Hopefully that’s something that continues to improve with initiatives like that of the HiPSTAS and evolving technology.
In response to your third question this reading has certainly made me more excited for the future of audio used in historical research. Although I am not particularly technically apt, Clement seems to have laid out quite well how to replicate the technique. I am also in the group with working with NPR and although as she has mentioned already there are other obstacles in the way of this kind of research, one of the main problems remains that having to listen to audio in order to determine its content is massively problematic when it comes to broader research topics. This work brings to mind Jockers’ Macroanalysis in the capabilities to review unreasonably large quantities of source material without the traditionally required man hours.