Oral Historians Have No Standards - At Least Around Transcription Style Guides
Michael Sesling, Audio Transcription Center
Hanassa Wicks, Audio Transcription Center
As a transcription service, we have a unique view of the plethora of style guides that exist among oral historians. The diverse landscape of style guides for transcribing oral histories presents a notable dichotomy. Our role allows us to offer a unique view, requiring us to delve into the minutiae of each style guide that we receive. In turn, this intricate process allows us a comprehensive understanding of the different approaches within the field.
ABSTRACT: As a transcription service, we have a unique view of the plethora of style guides that exist among oral historians. The diverse landscape of style guides for transcribing oral histories presents a notable dichotomy. Our role allows us to offer a unique view, requiring us to delve into the minutiae of each style guide that we receive. In turn, this intricate process allows us a comprehensive understanding of the different approaches within the field.In this session, we’d like to observe some of the various style guides that are used as a standard in the oral history world, and open a dialogue with the audience about all of the numerous and varied guidelines that exist.We’ll explore the potential of creating a more customizable style guideline for oral historians that utilize the centralized commonalities of the various guidelines used, and work with the audience to find a methodological approach to creating a more standardized concept that will help alleviate the challenges and difficulties encountered by transcriptionists, while remembering the nuances of each.Oral History Indexing (OHI)
Douglas Lambert, University at Buffalo, SUNY
In the Fall of 2023 I published an article in the Oral History Review called "Oral History Indexing (OHI)." In this presentation I will recap and comment on OHI's first 25-year innovation period, and discuss prospective new directions for this work in light of new technology opportunities.
ABSTRACT: Oral history indexing (OHI) is a set of practices that emerged for content management of large audio/video (A/V) collections. Driven by curators who wished to publish complete collections of interviews, and made possible by computer-based media, innovators in OHI introduced a variety of new modes of electronically linked A/V access, as early as the mid-1990’s. OHI typically involves creating thematic passages within recordings, with segments defined by media timecodes--providing access within and across interviews. Akin to an indexed book, OHI systems allow cross-referencing to specific points within media documents, describe content through natural language, and promote browsing and exploring modes rather than literal text searching. In this presentation I will summarize OHI between the 1990’s and 2023, when it existed as a set of markup processes facilitated by software tools but overseen by human intelligence. I will highlight a range of methodological approaches and system attributes from major institutions who pursued OHI and illustrate how the concepts and skills involved are applicable for other long-form A/V content. I will also discuss how emerging technologies such as automatic speech recognition and large language models are already changing oral history transcription practices, and OHI.The Warrior Women Project: Applying Ethics and Core Values in the Auditing and Editing Work of First Nations Oral History Transcripts
Eric Gaither, Columbia University
This session focuses on the praxis of oral history transcript auditors and editors who make an active commitment to honor nation and community-centered approaches and core values in the process of generating textual remembrances as artifacts.
ABSTRACT: The Warrior Women Project (WWP) is a collaborative of matriarchs, historians, community organizers, and multimedia storytellers working to bring to light the critical impact of Indigenous women throughout recent history. Over the past two and a half decades, WWP has generated thousands of hours of audiovisual content requiring some level of non-AI acknowledgement in the process of transcription. Multiple speakers interweaving remembrances across time, space, and language communities represent an embodied form of cultural practice that leads with the expectation an auditor and/or editor both hear and listen to narrators as expert, inter-legible community voice(s).This presentation focuses on auditory work that values deep, poly-vocal listening as a commitment to de-center self and center collective voice articulation within the aesthetic fabric of the oral history transcript. Rather than correct the voice in print, the approach employed here seeks to center and honor the voice; to approach the message and its speaker(s) with respect and humility; to engage in the work of translation from audio and visual to textual with integrity; and to acknowledge the wisdom of the matriarchs.Enhancing Oral History Collections in South Africa through Artificial Intelligence, Case Studies of Gauteng Unsung Sport Heroes and South African National Parks Board Conservation Records
Isabel Schellnack-Kelly, Department of Information Science, University of South Africa
Nampombe Saurombe, Department of Information Science, University of South Africa
The article focuses on trying to encourage two oral history collections to consider utilizing artificial intelligence tools to make their collections more accessible to archivists, researchers and interested members of the public. The challenges specifically relate to those encountered in South Africa.
ABSTRACT: Oral history collections serve as invaluable repositories of cultural heritage, offering insights into the lived experiences, perspectives, and narratives of individuals and communities. In South Africa, these collections play a crucial role in preserving and promoting the country's diverse heritage, yet they face significant challenges in management and accessibility, including labour-intensive transcription and indexing processes and the need for culturally sensitive approaches to archiving. In recent years, advances in artificial intelligence (AI) have provided new opportunities for enhancing the management and accessibility of oral history collections. By leveraging AI technologies such as speech recognition, natural language processing, and machine learning, archivists and researchers can automate tasks such as transcription, translation, and metadata tagging, thereby streamlining the archival process and expanding access to oral history materials. By undertaking a qualitative research within a postmodernist approach, this article explores the potential of AI in enhancing oral history collections in South Africa, with a focus on case studies of Gauteng unsung sport heroes and South African National Parks Board nature conservation records. Through these case studies, the researcher examines how AI technologies can address the challenges of managing and accessing oral history collections, while also considering the ethical implications and best practices for integrating AI into archival workflows.
