Automatic Speech Recognition on SlideSpeech S95 L95 (test)
8.46WERSLAM-LLM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SLAM-LLMTrain set=L95 (473h), Contextual Keywords=true, Adapter=LoRA2026.01 | 8.46 | 4.89 | 8.73 | 95.31 | |
| SLAM-LLMTrain set=L95 (473h), Contextual Keywords=false, Adapter=LoRA2026.01 | 8.61 | 7.34 | 8.72 | 92.84 | |
| SLAM-LLMTrain set=L95 (473h), Contextual Keywords=true2026.01 | 9.14 | 5.47 | 9.42 | 94.87 | |
| SLAM-LLMTrain set=L95 (473h), Contextual Keywords=false2026.01 | 9.34 | 9.52 | 9.33 | 90.64 | |
| SLAM-LLMTrain set=S95 (161h), Contextual Keywords=true2026.01 | 11.26 | 7.67 | 11.52 | 92.5 | |
| SLAM-LLMTrain set=S95 (161h), Contextual Keywords=false2026.01 | 11.8 | 13.52 | 11.71 | 86.71 | |
| LCB-netTrain set=L95 (473h), Contextual Keywords=true2026.01 | 12.02 | 9.03 | 12.24 | 91.12 | |
| CPPTrain set=L95 (473h), Contextual Keywords=true2026.01 | 12.38 | 9.32 | 12.6 | 90.86 | |
| SlidesSpeechTrain set=L95 (473h), Contextual Keywords=false2026.01 | 12.89 | 12.7 | 12.9 | 87.43 | |
| LCB-netTrain set=S95 (161h), Contextual Keywords=true2026.01 | 19.21 | 23.7 | 18.89 | 76.48 | |
| CPPTrain set=S95 (161h), Contextual Keywords=true2026.01 | 20.95 | 24.05 | 20.73 | 76.1 | |
| SlidesSpeechTrain set=S95 (161h), Contextual Keywords=false2026.01 | 21.22 | 26.6 | 20.83 | 73.51 |