Automatic Speech Recognition on English Multi-accent (evaluation set)
11.26WERSeed-ASR (ML)
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed-ASR (ML)Backbone=LUISE (2B parameters)2024.07 | 11.26 | |
| Universal-1Inference Type=API call2024.07 | 14.4 | |
| Whisper Large v3Inference Type=offline decoding2024.07 | 21.52 | |
| Google USMInference Type=API call2024.07 | 22.19 |