Multi-talker Automatic Speech Recognition on CH109
0.366Total WERGLAD-SOT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GLAD-SOTFine-tuning dataset=CH11-mix, Fine-tuning epochs=5, Number of parameters=35.31M2025.09 | 0.366 | 0.407 | |
| SOT+SACTC-13Fine-tuning dataset=CH11-mix, Fine-tuning epochs=5, Number of parameters=35.77M2025.09 | 0.374 | 0.419 | |
| CSE-SOTFine-tuning dataset=CH11-mix, Fine-tuning epochs=5, Number of parameters=38.90M2025.09 | 0.4 | 0.439 | |
| SOTFine-tuning dataset=CH11-mix, Fine-tuning epochs=5, Number of parameters=36.07M2025.09 | 0.454 | 0.468 |