UCLDH ONLINE: Audio/video thematic indexing | Historicizing modalities

Douglas Lambert, University of Buffalo, United States

Audio/video thematic indexing: meaning mapping for oral history access and usage

Audio/video timecode and thematic indexing is an area of technological and intellectual advancement that has been embraced in the oral history field in a variety of forms. The work developed from the technical ability to synchronize text or other metadata, via timecodes, to the long-form recorded sources they represent. Systems like OHMS (Oral History Metadata Synchronizer) display both transcripts and thematic indexes, both synchronized for direct and immediate listening to the related interview. Many collections, projects, and institutions favor thematic indexing by timecode for its flexibility and efficiency, versus focusing on word-for-word transcripts. Just as with a back of the book index, a thematic index for audio/video provides users an alternative to consuming a body of work linearly, and its entries are focused on content meaning, not just occurrences in the literal text.

Creating an index involves a balance between how one maps and structures the recorded content initially, optimizes timecode frequency and placement, and manages layers of metadata strategically. Beyond that, there can be seemingly unlimited possibilities for the markup to take shape, including higher order term management like controlled vocabularies organized in thematic maps (e.g., thesauri). I co-developed the Timecode Indexing Module (TIM) to experiment with alternate routes for populating the thematic index structure of OHMS and other a/v content display systems. TIM allows indexers to edit documents of variable text and timecode quality and to leverage improved automatic speech recognition (ASR) technology to develop new hybrid forms of timecoded text, making transcripts look more like indexes, and vice versa.

TIM was created by an international team, and the work was predicated on my long partnership with historian Michael Frisch at the University at Buffalo and The Randforce Associates, where we developed a variety of oral history and other a/v indexing methods with lay people, librarians, museum curators, and teams doing multidisciplinary qualitative research.

Alexander Freund, University of Winnipeg, Canada

Historicizing modalities: a few thoughts on oral history under surveillance capitalism

This presentation offers a few preliminary thoughts on doing oral history under the conditions of surveillance capitalism. It considers the political functions of transcriptions and archives; the interview as a “technology of the self” (e.g. confession, storytelling); and the commodification of personal data online. Historicizing our methods and tools—and the resulting modalities—is one way in which oral historians can reflect further on inherent assumptions and possibly unintended consequences of doing oral history in an environment that is increasingly shaped by unregulated global corporations that seek profits from commodifying personal data, personal time, and attention.

Multimodal Digital Oral History seminar series takes as its jumping off point, that the time is right to pursue a Multimodal Digital Oral History, or one that engages with oral history artefacts in all their representational modalities: transcript, sound, waveform, metadata and more. This seminar accordingly invites papers that explore any of the questions posed above, and in doing so contribute to the task of imagining a “Multimodal Digital Oral History” turn. The series will take place online (Zoom).

The joint virtual seminar series will be convened by Andrew Flinn (UCL) & Julianne Nyhan (TU Darmstadt & UCL) and co-hosted by the Centre for Critical Heritage Studies UCL; the Chair of Humanities Data Science and Methodology, TU Darmstadt, Germany; the International Centre for Archives and Records Management Research, UCL; and the UCL Centre for Digital Humanities.