Time Series On Speech Commands

% Test Accuracy

Results

Performance results of various models on this benchmark

		Paper Title
SepTr	98.51	SepTr: Separable Transformer for Audio Spectrogram Processing
ViT	98.11	AST: Audio Spectrogram Transformer
FlexTCN-4	97.73	FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes
MatchboxNet	97.40	MatchboxNet: 1D Time-Channel Separable Convolutional Neural Network Architecture for Speech Commands Recognition
CKCNN (100k)	95.27	CKConv: Continuous Kernel Convolution For Sequential Data
FlexTCN-6	-	FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizes

0 of 6 row(s) selected.