List your company
ResearchFormerly WizWand

Keyword Spotting on Speech Commands KS2 v2

98.9AccuracyASIT
86.83689.96893.196.232Apr 18, 2020Feb 25, 2021Jan 5, 2022Nov 15, 2022Sep 24, 2023Aug 3, 2024Jun 13, 2025
Updated 8mo ago

Evaluation results

MethodLinks
ASIT#Param=86M, Pre-training Data=AS2025.0698.9
A-JEPA#Param=86M, Pre-training Data=AS2025.0698.5
Audio-MAE#Param=86M, Pre-training Data=AS2025.0698.3
BEATSiter3#Param=90M, Pre-training Data=AS2025.0698.3
EAT#Param=88M, Pre-training Data=AS2025.0698.3
BEATSiter3+#Param=90M, Pre-training Data=AS, Supervised Fine-tuning=Included2025.0698.1
SSLAM#Param=88M, Pre-training Data=AS2025.0698.1
SS-AST#Param=89M, Pre-training Data=AS+LS2025.0698
MAE-AST#Param=86M, Pre-training Data=AS+LS2025.0697.9
SSASTModel size=base, Pretrained=true, Head learning rate=normal2024.0597.9
SSASTModel size=small, Pretrained=true, Head learning rate=larger2024.0597.8
MaskSpec#Param=86M, Pre-training Data=AS2025.0697.7
Embedding + Head# Parameters, K=3852020.0497.7
SSAMBAModel size=base, Pretrained=true, Head learning rate=normal2024.0597.4
SSASTModel size=tiny, Pretrained=true, Head learning rate=larger2024.0597.1
Harmonic Tensor 2D-CNN2020.0496.39
SSAMBAModel size=small, Pretrained=true, Head learning rate=larger2024.0596.3
SSAMBAModel size=base, Pretrained=false, Head learning rate=normal2024.0596.1
SSAMBAModel size=small, Pretrained=false, Head learning rate=larger2024.0595.7
Attention RNN# Parameters, K=2022020.0494.3
SSAMBAModel size=tiny, Pretrained=false, Head learning rate=larger2024.0594.2
SSAMBAModel size=tiny, Pretrained=true, Head learning rate=larger2024.0594
MSM-MAE#Param=86M, Pre-training Data=AS2025.0687.3