Automatic Speech Recognition on Mandarin Stuttered Speech (Moderate)
5.36CERQwen3-ASR-0.6B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-ASR-0.6BFine-tuned Dataset=AS-70 + Synthetic2026.06 | 5.36 | |
| Whisper-large-v3Fine-tuned Dataset=AS-70 + Synthetic2026.06 | 7.08 | |
| Qwen3-ASR-0.6BFine-tuned Dataset=AS-702026.06 | 7.51 | |
| Wav2Vec2.0Fine-tuned Dataset=AS-70 + Synthetic2026.06 | 9.33 | |
| Qwen3-ASR-0.6BFine-tuned Dataset=None2026.06 | 10.47 | |
| Whisper-large-v3Fine-tuned Dataset=AS-702026.06 | 12.49 | |
| Wav2Vec2.0Fine-tuned Dataset=AS-702026.06 | 13.12 | |
| Whisper-large-v3Fine-tuned Dataset=None2026.06 | 20.73 | |
| Wav2Vec2.0Fine-tuned Dataset=None2026.06 | 27.98 |