Speech Command Recognition on RAW Speech Commands (SC) Full 35 Labels (val)
96.78AccuracyLiquid-S4
Evaluation Results
| Method | Links | |
|---|---|---|
| Liquid-S4Parameters=224K, Sampling Rate=16kHz, p=22022.09 | 96.78 | |
| S4D-LinParameters=306K, Sampling Rate=16kHz2022.09 | 96.25 | |
| S4D-InvParameters=306K, Sampling Rate=16kHz2022.09 | 96.18 | |
| S4-LegSParameters=307K, Sampling Rate=16kHz2022.09 | 96.08 | |
| S4D-LegSParameters=306K, Sampling Rate=16kHz2022.09 | 95.83 | |
| ConvNetParameters=26.2M, Sampling Rate=16kHz2022.09 | 95.51 | |
| S4-(LegS/FouT)Parameters=307K, Sampling Rate=16kHz2022.09 | 95.32 | |
| S4-FouTParameters=307K, Sampling Rate=16kHz2022.09 | 95.27 | |
| S4D-InvParameters=306K, Sampling Rate=8kHz2022.09 | 91.8 | |
| S4-FouTParameters=307K, Sampling Rate=8kHz2022.09 | 91.59 | |
| S4D-LinParameters=306K, Sampling Rate=8kHz2022.09 | 91.58 | |
| S4-LegSParameters=307K, Sampling Rate=8kHz2022.09 | 91.32 | |
| S4D-LegSParameters=306K, Sampling Rate=8kHz2022.09 | 91.08 | |
| S4-(LegS/FouT)Parameters=307K, Sampling Rate=8kHz2022.09 | 90.72 | |
| Liquid-S4Parameters=224K, Sampling Rate=8kHz, p=22022.09 | 90 | |
| XResNet-50Parameters=904K, Sampling Rate=16kHz2022.09 | 83.01 | |
| ResNet-18Parameters=216K, Sampling Rate=16kHz2022.09 | 77.86 | |
| InceptionNetParameters=481K, Sampling Rate=16kHz2022.09 | 61.24 | |
| ResNet-18Parameters=216K, Sampling Rate=8kHz2022.09 | 8.74 | |
| XResNet-50Parameters=904K, Sampling Rate=8kHz2022.09 | 7.72 | |
| ConvNetParameters=26.2M, Sampling Rate=8kHz2022.09 | 7.26 | |
| InceptionNetParameters=481K, Sampling Rate=8kHz2022.09 | 5.18 |