Symposium: Annotation and Presentation of Multimodal Data
This symposium brings together researchers working on talk-in-interaction to address two interrelated questions: How can gesture annotation be further standardized to enhance comparability and reproducibility across studies? And what are effective strategies for representing the temporal, spatial, and sequential complexity of multimodal data in written and oral academic formats?
- https://www.uni-giessen.de/de/fbz/fb05/romanistik/aktuelles/cal/symposiom_multimodal
- Symposium: Annotation and Presentation of Multimodal Data
- 2026-06-16T13:30:00+02:00
- 2026-06-16T17:30:00+02:00
- This symposium brings together researchers working on talk-in-interaction to address two interrelated questions: How can gesture annotation be further standardized to enhance comparability and reproducibility across studies? And what are effective strategies for representing the temporal, spatial, and sequential complexity of multimodal data in written and oral academic formats?
16.06.2026 von 13:30 bis 17:30 (Europe/Berlin / UTC200)
Gustav-Krüger-Saal (Raum 105) Universitätshauptgebäude Ludwigstraße 23, 35390 Gießen or online
registration until June 08: registratiom form (link)
The analysis of multimodal interaction relies on detailed and systematic annotation of verbal and nonverbal resources, including gesture, gaze, body posture, and prosody. Yet despite growing consensus on the relevance of multimodal approaches, there is still considerable variation in how researchers annotate gestural and other bodily conduct — in terms of segmentation criteria, classification systems, annotation tools, and levels of granularity. At the same time, the question of how to effectively present multimodal data in academic publications and conference talks remains a practical and methodological challenge: static images, frame sequences, annotation screenshots, and dynamic formats each come with specific affordances and limitations.
This symposium brings together researchers working on talk-in-interaction to address two interrelated questions: How can gesture annotation be further standardized to enhance comparability and reproducibility across studies? And what are effective strategies for representing the temporal, spatial, and sequential complexity of multimodal data in written and oral academic formats? By fostering exchange on tools, conventions, and best practices, the symposium aims to contribute to a more transparent and interoperable methodology in multimodal interaction research.
|
Time (CEST) |
Presentation |
Format |
|
13.30-13.45h |
Welcome and greetings |
|
|
13.45-14.15h |
Prof. Dr. Ulrike Schröder (in person) / Giovanna Meneghin (online) (UFMG, Belo Horizonte): Transcribing Holistic Gestalts: Multimodal Conduct between Cognitive and Interactional Approaches |
hybrid |
|
14.15-14.45h |
Prof. Dr. Elwys de Stefani (University of Heidelberg): Transcribing bodies on the move |
in person |
|
14.45-15.15h |
Friederike Schulz (University of Potsdam): “Let the Pandas Sort Your Data”: A Python Pipeline for Quantitative Analysis of GAT-2 Transcripts |
in person |
|
15.15-15.45h |
Dr. Simone Gomes (UFMG) / Dr. Katharina Müller / Coralie Schindler (JLU Gießen): Automatic transcription of COVID-19 press conferences in Romance languages and annotation of gestures |
in person |
|
15.45-16.15h |
COFFEE BREAK |
|
|
16.15-16.45h |
Simon Titze (University of Heidelberg): The prosody and embodiment of assessments: what to transcribe and how |
online |
|
16.45-17.30h |
Discussion: Annotation and Presentation of Multimodal Data |
hybrid |