DARIAH’s Visual Media & Interactivity Working Group – Workshop on audiovisual corpora annotation [Oct 22nd, in person & online]

Workshop on audiovisual corpora annotation

October 22nd, 2025 [in person and online]

DARIAH’s Visual Media & Interactivity Working Group is participating in a one-day workshop on audiovisual corpora annotation, which will take place on October 22 during the 5th DARIAH-HR International Conference  (University of Osijek, Croatia). The workshop can also be followed remotely. Below is the link to the workshop presentation and to the registration form. 

https://dhh.dariah.hr/2025/workshops

The Consortium for Annotation, Analysis and Archiving of Video Applied to Scientific Activities (Canevas) is accredited since 2022 by the French research infrastructure for Digital Humanities Huma-Num. Since April 2025, this consortium has been receiving European EOSC-OSCARS funding for a period of 24 months, which has given rise to the OASIS project (Open Audiovisual Science Innovation Scheme). The aim of Canevas and OASIS is to facilitate research in the humanities and social sciences involving audiovisual corpora by facilitating actions such as archiving, annotating, commenting, analysing, and sharing videos. To do this, the members of the Canevas consortium have created two tools, Celluloid (for annotating corpora on media studies and media literacy) and e-spect@tor (for annotating corpora on the performing arts, especially theater), which enable collaborative annotation of videos for research or teaching purposes. These are free and open source tools (https://github.com/celluloid-camp/) that comply with open science and FAIR data standards, while leveraging AI to promote the intelligibility of videos and the interactions that result from them.

As part of the OASIS project, the Canevas Consortium is organising a workshop during the pre-conference day of the 5th DARIAH-HR International Conference, which will take place on Wednesday 22 October 2025. This workshop will be divided into two 3-hour sessions. The first will take place on Wednesday morning and aims to introduce participants to the PeerTube technology, developed by the French education-oriented video network Framasoft to offer an alternative to the services provided by the GAFAM, and particularly the online video hosting platform Youtube, and thus promote digital empowerment. The second session, on Wednesday afternoon, will be devoted to learning how to use the Celluloid and e-spect@tor tools. We invite you to discover collaborative annotation through your own audiovisual corpus, enabling you to develop new skills adapted to the changes of media practices and the epistemological issues that come with them. During this second session, we will focus on specific features provided by these tools: some are automated using AI, such as audio transcription and video segmentation into chapters, while others can be done manually and allow users to enrich their viewing experience through the traces they leave behind.

By renewing interactions and collaborations with these digital tools, this workshop aims to introduce participants, from all disciplines and all levels of video expertise, to our research methods while allowing to acquire new skills to foster a convergence culture around video archiving and annotations. These can then be deployed in various educational or research contexts, which can be enhanced, for instance, with group work in the classroom, or in research carried out by researchers in training (Masters, PhD) or more experienced researchers. 

The workshop will be led (in English) by:

  • Michael Bourgatte, Professor at the University of Lorraine (France)
  • Cécile Chantraine-Braillon, Professor at the University of La Rochelle (France)
  • Anatole Grimaldi, OASIS project engineer
  • Laurent Tessier, Professor at the Catholic University of Paris

NB: to get the most out of this workshop, please bring your own computer. It is also possible to follow this workshop with one computer for several people. Moreover, if you want to explore part of one of your corpora, you can send us one of your videos. All video formats are welcome.