Keyword Spotting on Google Speech Commands V2-35
98.11AccuracyAST-S
Evaluation Results
| Method | Links | |
|---|---|---|
| AST-SExtra Knowledge=Pre. ImageNet, #Params (M)=872021.10 | 98.11 | |
| AST-PExtra Knowledge=Pre. ImageNet & Audioset, #Params (M)=872021.10 | 97.88 | |
| KWT-2Extra Knowledge=KD with MHAtt-RNN, #Params (M)=2.3942021.10 | 97.74 | |
| KWT-3Extra Knowledge=KD with MHAtt-RNN, #Params (M)=5.3612021.10 | 97.69 | |
| KW-MLPExtra Knowledge=None, #Params (M)=0.4242021.10 | 97.56 | |
| KWT-2Extra Knowledge=None, #Params (M)=2.3942021.10 | 97.53 | |
| KWT-3Extra Knowledge=None, #Params (M)=5.3612021.10 | 97.51 | |
| KW-MLP10Extra Knowledge=KD with KW-MLP, #Params (M)=0.3532021.10 | 97.49 | |
| KW-MLP8Extra Knowledge=KD with KW-MLP, #Params (M)=0.2832021.10 | 97.45 | |
| KW-MLP10Extra Knowledge=None, #Params (M)=0.3532021.10 | 97.35 | |
| MHAtt-RNNExtra Knowledge=None, #Params (M)=0.7432021.10 | 97.27 | |
| KW-MLP8Extra Knowledge=None, #Params (M)=0.2832021.10 | 97.26 | |
| KW-MLP6Extra Knowledge=KD with KW-MLP, #Params (M)=0.2132021.10 | 97.17 | |
| KW-MLP6Extra Knowledge=None, #Params (M)=0.2132021.10 | 97.07 | |
| KWT-1Extra Knowledge=KD with MHAtt-RNN, #Params (M)=0.6072021.10 | 96.95 | |
| KWT-1Extra Knowledge=None, #Params (M)=0.6072021.10 | 96.85 | |
| Liquid-S4Parameters=224K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 96.78 | |
| S5Parameters=280K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 96.52 | |
| Res-15Extra Knowledge=None, #Params (M)=0.2372021.10 | 96.4 | |
| S4D-LinParameters=306K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 96.25 | |
| S4D-InvParameters=306K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 96.18 | |
| S4-LegSParameters=307K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 96.08 | |
| S4D-LegSParameters=306K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 95.83 | |
| ConvNetParameters=26.2M, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 95.51 | |
| S4-(LegS/FouT)Parameters=307K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 95.32 | |
| S4-FouTParameters=307K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 95.27 | |
| S5Parameters=280K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 94.53 | |
| Att-RNNExtra Knowledge=None, #Params (M)=0.2022021.10 | 93.9 | |
| S4D-InvParameters=306K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 91.8 | |
| S4-FouTParameters=307K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 91.59 | |
| S4D-LinParameters=306K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 91.58 | |
| S4-LegSParameters=307K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 91.32 | |
| S4D-LegSParameters=306K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 91.08 | |
| S4-(LegS/FouT)Parameters=307K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 90.72 | |
| Liquid-S4Parameters=224K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 90 | |
| XResNet-50Parameters=904K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 83.01 | |
| ResNet-18Parameters=216K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 77.86 | |
| InceptionNetParameters=481K, Sampling Rate=16000Hz, Zero-shot testing=false2022.08 | 61.24 | |
| ResNet-18Parameters=216K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 8.74 | |
| XResNet-50Parameters=904K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 7.72 | |
| ConvNetParameters=26.2M, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 7.26 | |
| InceptionNetParameters=481K, Sampling Rate=8000Hz, Zero-shot testing=true2022.08 | 5.18 |