Vosk + MediaPipe + LS + MixUp + SA + 3DResNet-18 + BiLSTM + Cosine WR
Reported on 2 benchmarks across 2 tasks
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Computer Vision1 result
- Top-1 Accuracy88.7best: 95 (SyncVSR (Word Boundary))
Natural Language Processing1 result
- Top-1 Accuracy88.7best: 95 (SyncVSR (Word Boundary))