List your company
ResearchFormerly WizWand

Audio-Visual Speech Recognition on LRS3 Noisy w.r.t Missing Modality 0.3 Masking Ratio (test)

5.77WER (%)M2S-AVSR
5.4067.86310.3212.777Jun 4, 2026
Updated 5mo ago

Evaluation results

MethodLinks
M2S-AVSRModality=A+V, Labeled Training Data (hrs)=ms1,759, Multi-view Synthesis=true2026.065.77
M2S-AVSRModality=A+V, Labeled Training Data (hrs)=1,7592026.065.84
MMS-LlamaModality=A+V, Labeled Training Data (hrs)=1,7592026.065.9
M2S-AVSRModality=A+V, Labeled Training Data (hrs)=ms433, Multi-view Synthesis=true2026.066.01
Auto-AVSRModality=A+V, Labeled Training Data (hrs)=3,4482026.066.11
MMS-LlamaModality=A+V, Labeled Training Data (hrs)=4332026.066.17
Llama-AVSRModality=A+V, Labeled Training Data (hrs)=1,7592026.066.75
M2S-AVSRModality=A+V, Labeled Training Data (hrs)=4332026.067.23
Whisper-FlamingoModality=A+V, Labeled Training Data (hrs)=1,7592026.068.22
Whisper-FlamingoModality=A+V, Labeled Training Data (hrs)=ms1,759, Multi-view Synthesis=true2026.068.42
Whisper-FlamingoModality=A+V, Labeled Training Data (hrs)=4332026.068.52
Whisper-FlamingoModality=A+V, Labeled Training Data (hrs)=ms433, Multi-view Synthesis=true2026.068.55
CMAModality=A+V, Labeled Training Data (hrs)=4332026.069.35
AV-RelScoreModality=A+V, Labeled Training Data (hrs)=4332026.0611.13
AV-HuBERTModality=A+V, Labeled Training Data (hrs)=4332026.0611.14
V-CAFEModality=A+V, Labeled Training Data (hrs)=4332026.0614.87