List your company

Audio-Visual Speech Recognition on LRS3 433 h 5 dB SNR

1.3WERMMS-LLaMA
1.2481.5991.952.301Mar 4, 2026

Evaluation results

MethodLinks
MMS-LLaMAModality=AV2026.031.3
AVUR-LLMModality=AV2026.031.4
AVGERModality=AV2026.031.9
Whisper-FlamingoModality=AV2026.032
Llama-AVSRModality=AV2026.032.2
CMAModality=AV2026.032.4
AV-HuBERTModality=AV2026.032.6