RIR-Mega是一个大规模的模拟房间脉冲响应数据集,旨在为机器学习和房间声学建模提供核心资源。该数据集包含大量的模拟房间脉冲响应,涵盖了广泛的房间几何形状和带限衰减特性。数据集使用紧凑的、易于解析的元数据架构,并提供了一系列实用工具,如元数据验证和校验和工具。RIR-Mega数据集通过Hugging Face Hub提供快速预览和流式传输,并在Zenodo上保存完整的50,000个房间脉冲响应存
There are two main folders: Source and Results. The Source folder contains two subfolders: Model Dataset and Simulation Dataset. The Model Dataset folder includes the original and BEM models, while th
This data set provides results from modeling of acoustic transmission loss in the ocean off central California, for a large-vessel sound source and a receiver on the continental slope.