Automatic Speech Recognition on Telephony
11.63WEREfficient Conformer
Evaluation Results
| Method | Links | |
|---|---|---|
| Efficient Conformerdecoding_mode=full-context2025.02 | 11.63 | |
| Squeezeformerdecoding_mode=full-context2025.02 | 11.72 | |
| ChunkFormercontext size [latt, c, r]=[256, 128, 128]2025.02 | 11.84 | |
| ChunkFormercontext size [latt, c, r]=[128, 256, 128]2025.02 | 11.85 | |
| ChunkFormercontext size [latt, c, r]=[128, 64, 128]2025.02 | 11.85 | |
| ChunkFormerdecoding_mode=full-context2025.02 | 11.86 | |
| Conformerdecoding_mode=full-context2025.02 | 12.34 |