By Silvia Calamai, Stef Scagliola, Henk van den Heuvel, Arjan van Hessen, Christoph Draxler
Who We Are
The project is a collaboration of a large, international group of scholars who met in Oxford in 2016 at a CLARIN-funded workshop. We are a group of experts interested in speech data with very different backgrounds – oral history, computational linguistics, anthropology, sociolinguistics, phonetics and phonology. We also share an interest in exploring how technology can be integrated into research that involves spoken narratives.
We started to envision ways of disseminating technology related to sound and transcription among the oral history community. The collaboration that started in Oxford led to a series of follow-up workshops: Utrecht 2017, Arezzo 2017, München 2018, Utrecht 2019, Sofia 2019 and Utrecht 2020. In order to give other researchers the opportunity to use the techniques, an oral history portal was created.
The Starting Point of the ‘Voices from Ravensbrück’ Project
The ‘Voices from Ravensbrück’ project began with the retrieval of the archive of Anna Maria Bruzzone, a writer who published the book Le donne di Ravensbrück in 1978, based on five interviews with survivors, which were recorded on cassettes. The story is fascinating. Just one year after Anna Maria Bruzzone passed away in 2015, Silvia Calamai initiated a search to trace audio cassettes with interviews conducted by Bruzzone for a completely different project (and book) on psychiatric patients of the former Arezzo psychiatric hospital (now the Siena University campus in Arezzo).
Eventually it turned out that they were in private hands, namely with the author’s niece, Paola Chiama, based in Turin. She generously lent the cassettes to Siena University so that they could be digitised and archived at the library. After having visited the campus and seen in person how the research on this topic by Silvia Calamai, Marica Setaro, Lucilla Gigli, Rosalba Nodari was progressing, she decided to donate her aunt’s entire oral archive, including the cassette tapes with interviews on Ravensbrück, to Siena University. The big question now was how to share this valuable material with the scholarly community, instead of simply digitising it and storing it in an archive?
The Project
The project ‘Voices from Ravensbrück’ aim to introduce a new type of corpus into the CLARIN Resource Family tree: a multilingual corpus of interviews curated and stored in a sustainable manner. The use of automatic speech recognition and alignment techniques will remove the need for time-consuming manual transcriptions. By using these technologies, the transcripts will also be enriched with extra features, such as pause and phoneme durations.
For Whom?
We envision that linguists, computer scientists and historians can all be attracted to the data, because the interviews cover a very compelling topic, and the way the data is presented will make it easier to take up and use.
At the same time, schools and associations can also use the data we are working on for dissemination activities and continuing education. For example, some excerpts from interviews were presented during the 2021 International Holocaust Remembrance Day, when students from Sienna University described their work on such data. They were deeply moved.
The Power of the Interviews
What is striking while listening to the interviews is the unbelievable strength of the interviewees – their solidarity in the camp, how they used their vernacular in order to say things that were forbidden, the relation between the ‘in-group’ and ‘the out-group’, and how their identity was transmitted by their language.
Moreover, there are some aspects that can only be accessed via the audio files: the questions raised by the interviewer, the overlapping of different speakers, the silences. In the edited book, there is no trace of the questions. In addition, it is difficult to get a sense of the amount of silences and the burden of silence.
Towards the Future
There are plans to extend the project to include other languages and the project team is currently exploring potential opportunities.
Thanks to:
Aldo Rolfi, Eligio Roveri, Giorgia Poli, Anna Maria and Paola Mori
Eleonora Bassi and Lucilla Gigli (Biblioteca Umanistica, Università degli studi di Siena)
Marica Setaro