|Henk van den Heuvel||Eric Sanders||Robin Rutten||Stef Scagliola||Paula Witkamp|
|CLST, RU, Nijmegen||CLST, RU, Nijmegen||CLST, RU, Nijmegen||Veteranen Instituut||DANS|
Interview data can be used in a number of ways, such as comparative research, restudy or follow-up study, re-analysis / secondary analysis, research design and methodological advancement, replication and validation of published work, and for teaching and learning.
Recent experiences with the re-use of interview data show that there is an enormous potential for this type of data in terms of multidisciplinary research, intersubjectivity and efficiency. Especially in the field of interview data related to the Second World War and other military conflicts several projects have successfully been carried out
In INTER-VIEWs the project partners (CLST, RU Nijmegen; DANS; Veterans Institute, Doorn; MPI, Nijmegen) have curated a corpus of 250 audio-interviews held with veterans of various conflicts , which can be searched through speech retrieval , and annotated and enriched by insights of the users through a so called search and annotation tool
The INTER-VIEWs project has provided the curation of about 250 audio-recorded life-history interviews with veterans, entailing:
- 120 World War II interviews which present a range of experiences of Dutch soldiers trained for and deployed in traditional warfare between 1935-1945.
- 100 interviews with veterans of the Dutch East Indies decolonization war between 1945 and 1950. This collection presents a large diversity of experiences at the local level in the domain of guerilla warfare.
- 30 interviews with veterans of the subsequent decolonization-war with Indonesia about the last colony New Guinea which ended in 1962. The interviews of this relatively unknown conflict present narratives about soldiers and local Papua-population left in uncertainty and isolation and the pressure of the international community to decolonize the area
Each interview lasts between 1.5 and 2.5 hours.
The interviews are stored permanently by DANS and will be accessible for the research community. The audio is stored in wav-format. All interviews contain extensive metadata offered in a standardized (CMDI) format. This will facilitate search queries by researchers through the CLARIN webportal.
Relevant metadata include (but are not restricted to):
- Interview subject
- Information about the mission or conflict
- Time period of the veteran’s mission or conflict
- Extensive Information about the interviewee
- Information about the interviewer
- Information about the audio, summary and transcripts
All interviews contain extensive textual summaries divided into time blocks of about 10 minutes, and topic keywords also arranged per 10 minutes. This information is stored in the metadata as well.
The annotation tool
In order to enlarge the search potential in the content of the interviews and to enrich these sources with the insights of the consulters a search and annotation tool has been developed .
|Webservice||Annotation Tool||Access to this Annotation Tool is arranged through a password protected website|
|Demo||Demonstration Version||Access to an Open Demonstration version of the tool is available|
|Documentation||Manual||The manual for the tool is availlable via the website|
All interviews are treated by using automatic speech recognition so that relevant keywords are spotted in the audio. Some 10 interviews have been manually transcribed.
By entering keywords the user of the tool can find relevant sections in the interview, view the metadata and link directly to the audio. He or she can further add annotations to selected fragments of the interviews and upload files (pictures etc.) to illustrate the annotation. These annotations can (but need not) be made visible to other researchers as well. In this way a wiki like working environment can be created containing a rich combination of raw interview data (audio, transcripts), metadata, and annotations. The project team is collecting funding to organize workshops for researchers of various disciplines in order to make them acquainted with this tool and to further tailor it to their needs.
The "Enriched Publication" based on this project, got a SURF-award in 2011, as can be seen on the photo.
- Van den Heuvel, H., Horik, R. van, Scagliola, S., Sanders, E., Witkamp, P. (2010): The VeteranTapes: Research Corpus, Fragment Processing Tool, and Enhanced Publications for the e-Humanities. In: Proceedings LREC2010, Malta.
- Van den Heuvel, H. (2010): Interviews, geknipt voor onderzoek. In:DIXIT 2010, jg 7 nr 1. pp. 20-21.
- Van den Heuvel, H., Sanders, E., Scagliola, S., Witkamp, P. (2011): Het INTER-VIEWs project. CLARIN NL boekje
- Van den Heuvel, H., Sanders, E., Rutten, R., Scagliola, S., Witkamp, P. (2012) An Oral History Annotation Tool for INTER-VIEWs. In Proceedings LREC2012, Istanbul, Turkey.