Table 1 The total number of trainable parameters and MACCPs of our proposed methods and several typical DNN-based RES methods
From: Nonlinear residual echo suppression based on dual-stream DPRNN
| Model | Model size | MACCPs |
|---|---|---|
| LSTM | 2.96M | 0.30G |
| MSTasNet | 14.8M | 23G |
| DSDPRNN_t | 2.84M | 22G |
| DSDPRNN_f | 2.77M | 22G |