针对声纹识别领域中基于传统Res2Net模型特征表达能力不足,泛化能力不强的问题,提出了一种结合稠密连接与残差连接的特征提取模块SE-DR-Res2Block(Sequeeze and excitation with dense and residual connected Res2Block).首先,介绍了应用传统Res2Block的ECAPA-TDNN(Emphasized channel attention,propagation and aggregation...