Our new X account is live! Follow @wizwand_team for updates

Visual Speech Recognition on LRS3

0.009WER

Auto-AVSR

Updated 4d ago

Evaluation Results

Method	Links
Auto-AVSR 2024.11		0.009
LP Conf 2024.11		0.009
Auto-AVSR 2024.11		0.01
USR 2024.11		0.011
ViT3D-CM 2024.11		0.016
RNN-T 2024.11		0.045
LP Conformer 2024.06		0.128
LP Conf 2024.11		0.128
LP Conf 2026.02		0.128
SynthVSR 2024.11		0.169
SynthVSR 2026.02		0.169
ViT 3D 2024.06		0.17
ViT3D-CM 2024.11		0.17
ViT3D-CM 2026.02		0.17
USR 2.0 2026.02		0.176
AutoAVSR 2024.06		0.191
Auto-AVSR 2024.11		0.191
Auto-AVSR 2026.02		0.191
BRAVEn w/ ST 2026.02		0.201
SyncVSR 2024.06		0.215
USR 2024.11		0.215
USR 2026.02		0.215
RAVEn-Large 2024.06		0.231
RAVEn w/ ST 2024.11		0.231
RAVEn w/ ST 2026.02		0.231
SyncVSR 2024.06		0.234
Auto-AVSR 2024.11		0.235
Auto-AVSR 2026.02		0.235
Llama-AVSR 2026.02		0.24
RAVEn-Large 2024.06		0.244
VATLM-Large 2024.06		0.262
AV-HUBERT-Large 2024.06		0.269
SyncVSR 2024.06		0.281
AV-HUBERT w/ ST 2024.11		0.286
AV-HuBERT w/ ST 2026.02		0.286
SyncVSR 2024.06		0.304
VTP 2024.06		0.307
VTP 2024.11		0.307
VTP 2026.02		0.307
SyncVSR 2024.06		0.312
CM-Aux 2024.06		0.315
AutoAVSR 2024.06		0.33
RAVEn 2024.06		0.331
SyncVSR 2024.06		0.333
RNN-T 2024.11		0.336
RNN-T 2026.02		0.336
VATLM 2024.06		0.342
CM-Aux 2024.06		0.347
AV-HUBERT 2024.06		0.348
AutoAVSR 2024.06		0.363
CM-Aux 2024.06		0.379
RAVEn 2024.06		0.391
VTP 2024.06		0.406
CM-Seq2Seq 2024.06		0.469
V2P 2024.11		0.551
V2P 2026.02		0.551
TM-Seq2Seq 2024.06		0.589
KD-Seq2Seq 2024.06		0.59
KD + CTC 2024.06		0.598