Automatic Speech Recognition on RT03 (test)
7.2SWB MetricSystem Combination (6+7+8+9) + TrafoXL LM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| System Combination (6+7+8+9) + TrafoXL LMTraining Data=SWB-2k, External LM=LSTM+TrafoXL2021.05 | 7.2 | 5.9 | 5.5 | 9 | |
| Single headed attention based sequence-to-sequence modelTraining Data=2k-lrg, External Language Model=true, Cross-utterance=true2020.01 | 8.2 | 6.4 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=2k-lrg, External Language Model=false, Cross-utterance=false2020.01 | 8.5 | 7 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=2k, External Language Model=true, Cross-utterance=false2020.01 | 9.8 | 6.7 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=2k, External Language Model=true, Cross-utterance=true2020.01 | 9.8 | 6.6 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=2k, External Language Model=false, Cross-utterance=false2020.01 | 10.4 | 7.2 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=300h, External Language Model=true, Cross-utterance=true2020.01 | 14.8 | 8.4 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=300h, External Language Model=true, Cross-utterance=false2020.01 | 15 | 8.8 | — | — | |
| Single headed attention based sequence-to-sequence modelTraining Data=300h, External Language Model=false, Cross-utterance=false2020.01 | 17.8 | 10.5 | — | — |