Speech Recognition on CallHome
12.24WERSeed-ASR (ML)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Seed-ASR (ML)Language=EN2024.07 | 12.24 | — | |
| ECLMCorrection model size=0.5B, Decoding strategy=greedy, Base model=Parakeet-TDT-v22024.05 | 13.5 | 13 | |
| Parakeet-TDT-v2Correction method=None2024.05 | 15.5 | — | |
| WhisperBackbone=Large V2, Evaluation Protocol=Zero-shot2022.12 | 17.6 | 49.4 | |
| Whisper Large-v2Language=EN2024.07 | 17.6 | — | |
| wav2vec 2.0Backbone=Large, Language Model=None2022.12 | 34.8 | — |