Automatic Speech Recognition on CMV-en
9.66WERWhisper-large-V3
Evaluation results
| Method | Links | |
|---|---|---|
| Whisper-large-V32026.06 | 9.66 | |
| Qwen-TTS-Tokenizer2026.06 | 10.4 | |
| S3-Tokenizer-FSQ2026.06 | 10.67 | |
| E2E-TTS-Stage2Training Stage=22026.06 | 14.53 | |
| E2E-TTS-Stage1Training Stage=12026.06 | 14.62 | |
| SenseVoice-small2026.06 | 14.71 | |
| w/o LLMAblation=Without LLM2026.06 | 14.77 | |
| w/o LLM and LFMAblation=Without LLM and LFM2026.06 | 16.53 |