MSFDnet:

Yi Chen1, Zhenyu Huang2, Liang Lei3

  • 1School of Big Data and Information Industry, Chongqing City Management College, No. 151, Daxuecheng South Second Road, Shapingba District, Chongqing 401331, China.

PubMed
概括

本研究介绍了MSDFnet,这是一种用于声音事件定位和检测 (SELD) 的新型模型,通过增强特征提取和融合来提高复杂音频的准确性. 新模型在动态场景中表现出色,性能优于现有方法.