Audio-Visual Speech Recognition on WildVSR (test)
0.385WERUSR 2.0
Evaluation Results
| Method | Links | |
|---|---|---|
| USR 2.0Labelled hours=656, Unlabelled hours=2,649, Lang. model=false, Shared params=true2026.02 | 0.385 | |
| Auto-AVSRLabelled hours=3,448, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.386 | |
| USRLabelled hours=433, Unlabelled hours=1,326, Lang. model=true, Shared params=true2026.02 | 0.464 | |
| RAVEn w/ STLabelled hours=433, Unlabelled hours=1,326, Lang. model=true, Shared params=false2026.02 | 0.467 | |
| AV-HuBERT w/ STLabelled hours=433, Unlabelled hours=1,326, Lang. model=true, Shared params=false2026.02 | 0.487 | |
| Auto-AVSRLabelled hours=1,759, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.493 | |
| AV-HuBERTLabelled hours=433, Unlabelled hours=1,326, Lang. model=false, Shared params=false2026.02 | 0.517 | |
| RAVEnLabelled hours=433, Unlabelled hours=1,326, Lang. model=false, Shared params=false2026.02 | 0.522 | |
| CM-seq2seqLabelled hours=1,459, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.584 | |
| Auto-AVSRLabelled hours=661, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.623 | |
| VTPLabelled hours=2,676, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.687 | |
| VTPLabelled hours=698, Unlabelled hours=0, Lang. model=true, Shared params=false2026.02 | 0.756 |