OMG-Empathy Dataset

OMG-Empathy数据集是由汉堡大学信息学院知识技术系等机构创建的，专注于测量故事对听众情感状态的影响。该数据集包含80个视频，总时长约7小时，记录了10名参与者与4名讲故事者之间的双人互动。每个视频中，讲故事者讲述一个故事，听众随后使用情感值尺度标注故事对其情感状态的影响。数据集旨在推动人工智能在共情和情感传染领域的应用，通过提供真实世界场景下的情感交互训练和评估。

The OMG-Empathy dataset was created by the Department of Knowledge Technologies, School of Informatics, University of Hamburg, and other institutions, focusing on measuring the impact of stories on listeners' emotional states. This dataset contains 80 videos with a total duration of approximately 7 hours, recording dyadic interactions between 10 participants and 4 storytellers. In each video, a storyteller tells a story, and the listener then annotates the impact of the story on their own emotional state using an emotional value rating scale. The dataset aims to promote the application of artificial intelligence in the fields of empathy and emotional contagion, by providing real-world scenario-based emotional interaction data for AI training and evaluation.