Automatic Speech Recognition on LibriSpeech other segmented (test)
3.9WEROurs.small
Evaluation results
| Method | Links | |
|---|---|---|
| Ours.smallDecoder=Joint CTC-Attention, Chunk Size (seconds)=3.842025.12 | 3.9 | |
| Ours.smallDecoder=Joint CTC-Attention, Chunk Size (seconds)=0.962025.12 | 4.4 | |
| Ours.smallDecoder=Attention Decoding, Chunk Size (seconds)=0.962025.12 | 4.6 | |
| Ours.smallDecoder=Attention Rescoring, Chunk Size (seconds)=0.962025.12 | 4.8 | |
| Ours.baseDecoder=Joint CTC-Attention, Chunk Size (seconds)=3.842025.12 | 4.9 | |
| Ours.baseDecoder=Joint CTC-Attention, Chunk Size (seconds)=0.962025.12 | 5.4 | |
| Ours.baseDecoder=Attention Decoding, Chunk Size (seconds)=0.962025.12 | 5.5 | |
| Ours.baseDecoder=Attention Rescoring, Chunk Size (seconds)=0.962025.12 | 5.9 | |
| Whisper small.enDecoder=Attention Decoding, Chunk Size (seconds)=302025.12 | 6.7 | |
| Whisper base.enDecoder=Attention Decoding, Chunk Size (seconds)=302025.12 | 9.6 |