Chinese Multimodal Depression Corpus

This dataset contains semi-structural interviews designed to support the screening and assessment of major depressive disorder in China. These interviews were collected as part of a larger effort to create automatic AI tools that interview people and identify visual, acoustic, and textual indicators of MDD. Audio-visual recordings of interviews, extensive questionnaire responses are collected and have been transcribed and annotated for a variety of visual, acoustic, and textual features. The package includes 78 folders for each subject. All subjects were audio recorded, and only 45 subjects (19 MDD and 26 HC) were both audio and video recorded.

本数据集包含用于辅助中国人群重性抑郁障碍（Major Depressive Disorder，MDD）筛查与评估的半结构化访谈内容。该类访谈属于一项大型研究项目的组成部分，该项目旨在开发可开展人机访谈并识别MDD相关视觉、声学及文本指征的自动AI智能体工具。研究团队采集了访谈的音视频记录与大量问卷应答数据，并针对各类视觉、声学及文本特征完成了转录与标注工作。本数据集包包含78个对应每名受试者的文件夹。所有受试者均接受了音频录制，其中仅45名受试者（19名MDD患者与26名健康对照）同时完成了音视频录制。

搜集汇总

背景与挑战

背景概述

Chinese Multimodal Depression Corpus是一个用于支持中国重度抑郁症筛查和评估的多模态数据集，包含78个受试者的半结构化访谈音频和视频记录，以及问卷回答，已标注多种视觉、听觉和文本特征。其中45个受试者（19个MDD和26个HC）有音频和视频记录。