List your company
ResearchFormerly WizWand

Visual Speech Recognition on LRS3 low-resource (test)

19.3WERSerdyuk et al. (2022)
16.25236.82657.477.974Dec 12, 2022
Updated 8mo ago

Evaluation results

MethodLinks
Serdyuk et al. (2022)Encoder=Conformer, LM=false, Lab hrs=90,000*2022.1219.3
RAVEn w/ self-trainingEncoder=Transformer, LM=true, Unlab hrs=1,759, Lab hrs=30, Model size=Large2022.1223.8
RAVEn w/ self-trainingEncoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Large2022.1224.8
Serdyuk et al. (2021)Encoder=Transformer, LM=false, Lab hrs=90,000*2022.1225.9
Shi et al. (2022) w/ self-trainingEncoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Large2022.1228.6
Prajwal et al. (2022)Encoder=Transformer, LM=true, Lab hrs=2,676*2022.1230.7
Shi et al. (2022)Encoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Large2022.1232.5
RAVEnEncoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Large2022.1232.5
Makino et al. (2019)Encoder=RNN, LM=false, Lab hrs=31,000*2022.1233.6
Ma et al. (2022)Encoder=Conformer, LM=true, Lab hrs=8132022.1234.7
RAVEnEncoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Base2022.1240.2
Shi et al. (2022)Encoder=Transformer, LM=false, Unlab hrs=1,759, Lab hrs=30, Model size=Base2022.1246.1
RAVEnEncoder=Transformer, LM=false, Unlab hrs=433, Lab hrs=30, Model size=Base2022.1247
Shi et al. (2022)Encoder=Transformer, LM=false, Unlab hrs=433, Lab hrs=30, Model size=Base2022.1251.8
Shillingford et al. (2019)Encoder=RNN, Lab hrs=3,886*2022.1255.1
Xu et al. (2020)Encoder=RNN, LM=false, Lab hrs=5902022.1257.8
Afouras et al. (2018a)Encoder=Transformer, LM=true, Lab hrs=1,519*2022.1258.9
Zhang et al. (2022)Encoder=Transformer, LM=false, Unlab hrs=433, Lab hrs=30, Model size=Base2022.1267.8
Base modelEncoder=Transformer, LM=false, Lab hrs=30, Pre-training=from scratch2022.1293.4
Large modelEncoder=Transformer, LM=false, Lab hrs=30, Pre-training=from scratch2022.1295.5