List your company
ResearchFormerly WizWand

Speech Classification on Speech Commands v1

99AccuracyAudioMosaic
5.08829.46953.8578.231May 14, 2026May 17, 2026May 21, 2026May 24, 2026May 28, 2026May 31, 2026Jun 4, 2026
Updated 4mo ago

Evaluation results

MethodLinks
AudioMosaicBackbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=86M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0599
SSLAMBackbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=93M, Fine-tuning Parameters=88M, Evaluation Protocol=fine-tuning2026.0598.8
ASiTBackbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=96M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0598.2
BEATsiter3Backbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=182M, Fine-tuning Parameters=91M, Evaluation Protocol=fine-tuning2026.0597.7
A-JEPABackbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=354M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0597.7
Audio-MAEBackbone=ViT-B/16, Pre-training Data=AS, Pre-training Parameters=137M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0596.9
F3-TokenizerProbing=Frozen representation2026.0696.8
HuBERTBackbone=ViT-B/16, Pre-training Data=LS, Pre-training Parameters=95M, Fine-tuning Parameters=95M, Evaluation Protocol=linear evaluation2026.0596.3
Wav2Vec 2.0Backbone=ViT-B/16, Pre-training Data=LS, Pre-training Parameters=95M, Fine-tuning Parameters=95M, Evaluation Protocol=linear evaluation2026.0596.2
Ming-UProbing=Frozen representation2026.0696.14
SS-ASTBackbone=ViT-B/16, Pre-training Data=AS+LS, Pre-training Parameters=89M, Fine-tuning Parameters=89M, Evaluation Protocol=fine-tuning2026.0596
MAE-ASTBackbone=ViT-B/16, Pre-training Data=AS+LS, Pre-training Parameters=174M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0595.8
ASTBackbone=ViT-B/16, Pre-training Data=IN, Pre-training Parameters=86M, Fine-tuning Parameters=86M, Evaluation Protocol=fine-tuning2026.0595.5
WhisperProbing=Frozen representation2026.0693.3
BYOL-ABackbone=CNN, Pre-training Data=AS, Pre-training Parameters=5M, Fine-tuning Parameters=5M, Evaluation Protocol=fine-tuning2026.0591
w/o RQProbing=Frozen representation2026.0688.8
w/o LLMProbing=Frozen representation2026.0684.6
COLABackbone=CNN, Pre-training Data=AS, Pre-training Parameters=5M, Fine-tuning Parameters=5M, Evaluation Protocol=fine-tuning2026.0576.7
No repr.Probing=Frozen representation2026.068.7