Live respeaking - VRT case study

Erik de Snerck presented VRT's many years of experience in the creation of multilingual live subtitling by respeaking. In order to meet VRT’s claim "relevant to everybody", the challenge is to break down barriers that also arise from multilingualism and to keep the delay as low as possible by delaying the program.

Read More
CompAsS Team
From Audio Mining to Video Mining

Dr.-Ing. Joachim Köhler from Fraunhofer IAIS presented the development of the speech recognition tool working with neural networks and leading to audio mining and followed by an axplanation of video mining, the recognition of images, with an example from practice, namely face recognition.

Read More
CompAsS Team
Subtitles for immersive media

Ronald Mies of Institut für Rundfunktechnik deepened in his lecture the challenges of subtitling in 360° space and listed several approaches to positioning, among them the always visible subtitles or the speaker-fixed subtitles, which however require an indication of the position of the speaker. For this position display several models are tested and evaluated.

Read More
CompAsS Team
SUBTITLES AS A CHANCE FOR EUROPE

Amélie Leenhardt of ARTE G.E.I.E. presented the EU-funded ARTE Europe project, which aims to make ARTE content available in English, Polish, Italian and Spanish as well as in Germany and France through foreign language subtitles. Chloé Roux of ARTE G.E.I.E. presented the production of language versions at ARTE, which in addition to subtitling also includes the production of voice-over versions, dubbing and translations.

Read More
CompAsS Team
Subtitling and Translation Process Research

Within CompAsS scientific support is provided by a team from the Translation & Cognition Center (TraCo) at the Johannes Gutenberg University Mainz. This research group has as its central aim the cognitive modelling of the translation process with the help of empircal research methods such as corpora, keylogging, eytracking and EEG.

Read More
CompAsS Team