What type of study is this?

This is a Quantitative Study study.

September 29, 2025Open Access

StandUp4AI: A New Multilingual Dataset for Humor Detection in Stand-up Comedy Videos

View Full Paper

Ask AI

Bookmark

Authors

This multilingual dataset improves humor detection in stand-up comedy, suggesting a new labeling approach to enhance laughter detection.

This dataset includes over 330 hours of stand-up comedy across seven languages, aimed at improving humor detection.
The dataset is annotated for audience laughter, with a portion manually labeled for validation to support effective training.
This approach frames humor detection as word-level sequence labeling, allowing for more context-aware detection mechanisms.
The study proposes enhancements to automatic laughter detection through improvements in Audio Speech Recognition error handling.

Barrière et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark