REMIT

Summary

In the rapidly evolving landscape of media and technology, the way we interact with and consume media content is undergoing a significant transformation. Furthermore, the question of content veracity is an issue we already have to face, in order to find attractive and reliable information solutions.

What if the way we rapidly produce news became an opportunity? Indeed, this “fast fashion of news” engages a huge amount of single-use information, then stored and categorized as archives, whether it is the latest TV news or a documentary made in 1980.

REMIT is a project developing an innovative system that leverages cutting-edge AI technologies to create user-driven narratives from the expansive video archive of Radio Télévision Suisse (RTS). Too technical? Then imagine an immersive system based on 360° screens where people will have the opportunity to ask and converse with an AI that will process archives, and offering a new narrative around the discussed subject.

Keywords :
archives, museology, interactive, narratives, LLM
Original title

“Revolutionizing Media Interaction: A Novel AI System for Dynamic Narratives from Audiovisual Archives (REMIT)”

Duration

24 months

People involved

Principal investigator

Sarah Kenderdine (Lead Laboratory for Experimental Museology, EPFL)

 

Scientists

André Rattinger (Software Engineer, EPFL)

Loïc Serafin (3D Engineer, EPFL)

Kirell Benzi (Tech Lead, EPFL)

 

Media expert

Léonard Bouchet (Head of Data and Archives, RTS)

Academic Institution

Laboratory for Experimental Museology – eM+ (EPFL)

Media Partner

RTS – Radio Télévision Suisse (SRG) – Data and Archives (D+A)

Status

This project started in April 2024 and is ongoing

Related call for projects

Shaping the future of media and society: multidisciplinary projects on immersive and connected experiences and the open web” (2023)