D3BV

D3BV是一个专注于鸟类方言的跨语料库鸟类物种识别数据集，由北京林业大学技术学院和慕尼黑工业大学健康信息学主席等机构合作创建。该数据集包含超过25小时的音频记录，涵盖美国大陆三个不同地区的10种鸟类。数据集的创建旨在解决鸟类识别中由于方言差异带来的挑战，特别是在不同地理区域的鸟类声音识别。D3BV通过详细标注和预处理，为研究者提供了一个高质量的研究平台，以探索和改进跨地区鸟类识别技术。

D3BV is a cross-corpus bird species identification dataset focused on avian dialects, co-developed by institutions including the School of Technology at Beijing Forestry University and the Chair of Health Informatics at the Technical University of Munich, among others. This dataset contains over 25 hours of audio recordings covering 10 bird species across three distinct regions of the continental United States. It was created to address the challenges posed by dialectal differences in bird recognition, particularly for acoustic bird identification across different geographic regions. Through detailed annotation and preprocessing, D3BV provides researchers with a high-quality research platform for exploring and improving cross-regional bird recognition technologies.