基于卷积和门控注意的两阶段视听语音增强算法

王盼蓉,贾海蓉,段淑斐

Two-stage audio-visual speech enhancement algorithm based on convolution and gated attention

Panrong WANG,Hairong JIA,Shufei DUAN

表 1 基于卷积和门控注意的两阶段视听语音增强算法的实验参数设置

Tab.1 Experimental parameter setting for two-stage audio-visual speech enhancement algorithm based on convolution and gated attention

参数 数值 Conv1D卷积核大小L 16 块长度P 80 双阶段CNN-GAU模块个数R 8 ${{\boldsymbol{X}}'_{\text{U}}}$、${{\boldsymbol{X}}'_{\text{I}}}$、${{\boldsymbol{V}}_{\text{U}}}$特征维度M 512 ${{\boldsymbol{X}}'_{\text{Z}}}$、${{\boldsymbol{V}}_{\text{Z}}}$特征维度D 128