| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multi-speaker Automatic Speech Recognition | Aishell4 (eval) | Character Error Rate (CER)17.17 | 13 | |
| Speaker Attribute Prediction | AISHELL4 Eval (test) | Accuracy (ACC)97.12 | 3 | |
| Speaker-attributed Automatic Speech Recognition | AISHELL4 Long-form (test) | DER15.32 | 2 |