Speaker-attributed Automatic Speech Recognition on AMI SDM
18.6cpWERDiCoW v3.3
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DiCoW v3.3Model Category=Specialized MT-ASR Models, Diarization System=DiariZen2026.06 | 18.6 | — | — | — | |
| DixtralModel Category=General Purpose Spoken LMs, Diarization System=DiariZen2026.06 | 19.8 | — | — | — | |
| GLSC-SDR2026.03 | 23.32 | 17.49 | 5.83 | 76.67 | |
| Gemini-3-pro*source=converted from [18]2026.03 | 26.91 | 22.09 | 4.82 | — | |
| Qwen2.5-omni-sftmode=fine-tuned2026.03 | 27.16 | 20.23 | 6.93 | 74.33 | |
| ViebVoice-ASR2026.03 | 28.82 | 24.65 | 4.17 | — | |
| VibeVoiceModel Category=Specialized MT-ASR Models2026.06 | 33.7 | — | — | — | |
| Gemini-2.5-pro*source=converted from [18]2026.03 | 34.78 | 22.35 | 12.43 | — | |
| Voxtral MTv2Model Category=Specialized MT-ASR Models2026.06 | 42.3 | — | — | — | |
| TagSpeech2026.03 | 42.55 | 31.62 | 10.93 | 70.01 | |
| Qwen2.5-omni2026.03 | 49.86 | 32.25 | 17.61 | 60.06 | |
| Gemini 3.0 FlashModel Category=General Purpose Spoken LMs2026.06 | 56.3 | — | — | — |