[2312.08603] NeXt-TDNN: Modernizing Multi-Scale Temporal …
https://arxiv.org/abs/2312.08603
WEBDec 14, 2023 · NeXt-TDNN: Modernizing Multi-Scale Temporal Convolution Backbone for Speaker Verification. In speaker verification, ECAPA-TDNN has shown remarkable improvement by utilizing one-dimensional (1D) Res2Net block and squeeze-and-excitation (SE) module, along with multi-layer feature aggregation (MFA).
DA: 74 PA: 67 MOZ Rank: 84