Skip to main content

Table 2 Performance of our proposed methods and several typical RES methods

From: Nonlinear residual echo suppression based on dual-stream DPRNN

Echo Model PESQ SDR STOI
Artificial speech LAEC 1.48 −2.60 0.622
  LSTM 2.14 6.33 0.780
  MSTasNet 2.54 11.6 0.857
  DSDPRNN_ty 2.61 12.3 0.866
  DSDPRNN_tx 2.66 12.8 0.876
  DSDPRNN_fy 2.75 12.4 0.880
  DSDPRNN_fx 2.74 12.5 0.882
Artificial music LAEC 1.48 −2.90 0.634
  LSTM 2.08 5.46 0.755
  MSTasNet 2.43 10.7 0.830
  DSDPRNN_ty 2.50 11.5 0.842
  DSDPRNN_tx 2.61 12.6 0.865
  DSDPRNN_fy 2.62 11.4 0.857
  DSDPRNN_fx 2.64 11.6 0.863
ER speech LAEC 16.1 −2.05 0.697
  LSTM 2.13 4.85 0.799
  MSTasNet 2.66 11.6 0.890
  DSDPRNN_ty 2.68 11.7 0.892
  DSDPRNN_tx 2.62 11.5 0.887
  DSDPRNN_fy 2.77 11.3 0.904
  DSDPRNN_fx 2.66 10.6 0.895
ER music LAEC 1.70 −1.12 0.730
  LSTM 2.25 5.95 0.826
  MSTasNet 2.72 12.2 0.898
  DSDPRNN_ty 2.75 12.6 0.900
  DSDPRNN_tx 2.68 12.3 0.897
  DSDPRNN_fy 2.79 11.9 0.907
  DSDPRNN_fx 2.76 11.7 0.907
LL speech LAEC 1.95 1.67 0.806
  LSTM 2.55 9.23 0.884
  MSTasNet 2.99 15.0 0.932
  DSDPRNN_ty 3.00 15.6 0.932
  DSDPRNN_tx 2.87 14.9 0.920
  DSDPRNN_fy 3.02 15.3 0.938
  DSDPRNN_fx 3.04 15.7 0.938
LL music LAEC 1.97 2.16 0.820
  LSTM 2.60 9.07 0.889
  MSTasNet 3.04 15.6 0.934
  DSDPRNN_ty 3.07 16.0 0.935
  DSDPRNN_tx 2.89 14.8 0.921
  DSDPRNN_fy 3.12 15.8 0.944
  DSDPRNN_fx 3.13 16.0 0.943