GaMMA Corpus: Polyadic Conversations in Danish with Gaze, Speech, and Motion Data in Quiet and Adverse Speech Conditions

The GaMMA (Gaze, Motion, and Multi-talker Audio) corpus captures the behavior of polyadic conversations among native Danish speakers under both normal and cocktail party conditions. Eleven groups of four normal-hearing participants (44 unique individuals) are recorded while engaged in natural and spontaneous interactions. All conversations were conducted without conversational tasks. Each group was intentionally composed of participants with prior intragroup and interpersonal relations. Gaze and motion data were collected using an optical tracking system (Vicon) and eye-tracking glasses (Tobii), while speech was recorded via omnidirectional head-worn microphones and binaural hearing aid microphones with low occlusion. Calibrations were conducted before trials and compensation filters were created to account for differences in microphone placements. Processed versions of the audio signals, with background noise attenuated and crosstalk removed, were used to compute speech activity for all participants. The corpus, including both raw and processed gaze and audio data, as well as filters, calibration signals, and speech activity output, is included in this dataset.The Data descriptor for the corpus titled "The GaMMA corpus of Danish polyadic conversations with gaze speech and motion data in quiet and noise" can be found via this DOI: (will be added once journal finishes final review).Full annotations for the corpus are available at: https://doi.org/10.5061/dryad.r7sqv9snc (in review, DOI not active yet)and described in this article: https://aclanthology.org/2025.sigdial-1.19/GaMMA（Gaze, Motion, and Multi-talker Audio，注视、动作与多说话者音频）语料库收录了以丹麦语为母语的说话者在常规环境与鸡尾酒会场景下的多人间会话行为数据。本研究共录制11组受试参与者，每组包含4名听力正常的个体（总计44名独立参与者），受试过程中参与者开展自然自发的互动会话。所有会话均未设置特定会话任务，且每组参与者均由预先存在组内及人际关联的人员组成。 研究人员通过光学追踪系统（Vicon）与眼动追踪眼镜（Tobii）采集注视与动作数据，同时通过全向头戴式麦克风及低封堵式双耳助听器麦克风录制语音信号。 实验正式开展前需完成校准流程，并针对麦克风摆放位置差异制作补偿滤波器。研究人员将经背景噪声衰减、串音去除处理后的音频信号版本，用于计算所有参与者的语音活动状态。 本数据集包含该语料库的全部内容，涵盖原始与经处理的注视、音频数据，以及滤波器、校准信号与语音活动输出结果。该语料库的数据集说明文档标题为《GaMMA语料库：安静与噪声环境下丹麦语多人间会话的注视、语音与动作数据》，其DOI将在期刊完成最终审核后补充。 该语料库的完整标注数据可通过以下链接获取：https://doi.org/10.5061/dryad.r7sqv9snc（当前处于审核阶段，DOI尚未正式生效），相关描述详见论文：https://aclanthology.org/2025.sigdial-1.19/