KU Leuven audiovisual, gaze-controlled auditory attention decoding (AV-GC-AAD) dataset
收藏资源简介:
KU Leuven视听觉注意力解码(AV-GC-AAD)数据集由鲁汶大学创建,旨在研究在不同视听条件下,参与者对两个竞争说话者的注意力方向。数据集包含16名正常听力参与者的脑电图(EEG)信号,记录了他们在不同视觉条件下的注意力反应。数据集分为4种条件,每种条件有2个10分钟的试验,总计8个试验。数据集的创建旨在揭示现有空间注意力解码算法中的视觉注意力偏差,并提供一个基准用于未来的注意力解码算法评估。该数据集的应用领域主要集中在认知控制助听器和空间注意力解码算法的改进。
The KU Leuven Auditory-Visual Attention Decoding (AV-GC-AAD) Dataset was developed by KU Leuven to investigate participants' attentional orientation towards two competing speakers under diverse audiovisual conditions. It comprises electroencephalogram (EEG) signals from 16 normal-hearing participants, capturing their attentional responses across varying visual conditions. The dataset is categorized into 4 conditions, with two 10-minute trials per condition, totaling 8 trials in all. The core objective of this dataset is to uncover visual attentional biases existing in current spatial attention decoding algorithms, and to provide a benchmark for evaluating future attention decoding algorithms. Its main application areas focus on cognitive hearing aids and the improvement of spatial attention decoding algorithms.




