Open.Oral-History: Recommendations and tools for risk assessment, anonymization, and the provision of legally protected and ethically sensitive audiovisual interviews
Oral history's narrative-biographical interviews present specific ethical and legal protection requirements. These results in significant hurdles to their availability. At the same time, however, there is considerable interdisciplinary interest in reusing them. The "Open.Oral-History" project addresses this need by helping collection managers systematically assess the prerequisites for making their interviews available. Where necessary, the project enables this through an AI-supported anonymization process.
Many collection managers lack structured knowledge of the legal conditions and technical possibilities for digital provision. Complex restrictions related to data protection, privacy, and copyright—such as missing or insufficient consent forms or the required anonymization—often seem impossible to overcome. In collaboration with owners of interview collections we identify exemplary use cases. With legal support, workflows for legal clarification and risk assessment are developed for these various scenarios. The results will be published as a guide.
Oral history interviews are more difficult to anonymize than other research data because they contain both audio and visual information in the form of moving media. Anonymization can range from obscuring specific audio or visual elements to preventing the re-identification of individuals. At the same time, to maintain the integrity of qualitative research, the source should be altered as little as possible. Machine-readable transcripts are automatically generated using AI-based speech recognition. Words and passages to be anonymized are identified via named entity recognition, and then automatically obscured in the audio and/or video following a qualitative selection. Collection managers receive anonymized text derivatives, audiovisual media, and machine-readable license and rights information for providing and reusing the oral history interviews.
The legally reviewed guidelines and the anonymization workflow developed as a prototype are intended to be applicable to datasets from other academic disciplines. The transferability of the developed solutions will be determined through collaboration with other pilot projects and representatives of related disciplines.
Duration:01.03.2025 - 28.02.2027
Funding:The project is funded by the German Research Foundation (DFG) as part of the pilot phase “Digitization and Provision of (Currently) Copyright-Protected Objects” within the “Digitization and Access” funding program.
Links:- (German language) Article about the project on the blog "Rechtebewehrte Sammlungsobjekte": https://schutzrechte.hypotheses.org/1206
Freie Universität Berlin, Digitale Interview Collections
- Dr. Dennis Mischke (Submission of Application)
- Verena Lucia Nägel (Project Lead)
- Dr. Tobias Kilgus (Project Lead)
- Peter Kompiel
- Marc Altmann
- Dr. Almut Leh ((Submission of Application/ Project Lead)
- Dr. Dennis Möbus
- Phillip Bayerschmidt
- Helmut Hofbauer
Freie Universität Berlin
University Library
Digitale Interview Collections
Project: Open.Oral-History
Garystraße 39 | 14195 Berlin
Tel.: +49 30 838-52533
Email: interviewsammlungen@ub.fu-berlin.de
Website: https://www.fu-berlin.de/ooh

