LIPSFUS

LIPSFUS是一个专为音频-视觉感官融合设计的神经形态数据集，由西班牙塞维利亚大学的计算机与机器人技术实验室创建。该数据集包含26,620条记录，涵盖多种语言处理相关的词汇和短句，如数字、机器人命令等。数据集通过精确的时间同步技术，使用地址事件表示传感器和工具收集。创建过程中，数据集从不同国籍和年龄的人群中收集，确保了多样性。LIPSFUS数据集适用于基于人工和脉冲神经网络算法的感官融合架构，旨在解决机器学习应用中的唇读问题。

LIPSFUS is a neuromorphic dataset dedicated to audio-visual sensory fusion, developed by the Computer and Robotics Technology Laboratory at the University of Seville, Spain. This dataset comprises 26,620 records, including vocabulary and short sentences related to various language processing tasks such as numbers, robot commands, and the like. It was collected using Address Event Representation (AER) sensors and supporting tools, with precise time synchronization implemented throughout the data acquisition process. During its construction, the dataset was gathered from participants across different nationalities and age groups to ensure data diversity. The LIPSFUS dataset is applicable to sensory fusion architectures based on artificial neural network (ANN) and spiking neural network (SNN) algorithms, and is designed to address the lip-reading challenge in machine learning applications.