Automatic Speech Recognition on Reading
5WERConformer
Evaluation Results
| Method | Links | |
|---|---|---|
| Conformerdecoding_mode=full-context2025.02 | 5 | |
| ChunkFormercontext size [latt, c, r]=[128, 64, 128]2025.02 | 5.05 | |
| ChunkFormerdecoding_mode=full-context2025.02 | 5.08 | |
| ChunkFormercontext size [latt, c, r]=[128, 256, 128]2025.02 | 5.11 | |
| ChunkFormercontext size [latt, c, r]=[256, 128, 128]2025.02 | 5.13 | |
| Squeezeformerdecoding_mode=full-context2025.02 | 5.2 | |
| Efficient Conformerdecoding_mode=full-context2025.02 | 5.25 |