Automatic Speech Recognition on TED-LIUM
0.514WERUnited-MedASR
Evaluation results
| Method | Links | |
|---|---|---|
| United-MedASR2024.11 | 0.514 | |
| Qwen3-OmniOpen-Source=true, Size=30B-A3B, Reasoning mode=Reasoning off2026.04 | 2.4 | |
| DSM-ASR2026.03 | 2.9 | |
| Nemotron 3 Nano OmniOpen-Source=true, Size=30B-A3B, Reasoning mode=Reasoning off2026.04 | 3.11 | |
| CrisperWhisperSize/h=10,000, Params/B=1.50, Evaluation Protocol=short-form2026.01 | 3.2 | |
| CanarySize/h=31,200, Params/B=1.00, Evaluation Protocol=short-form2026.01 | 3.56 | |
| TRADEdecoding=streaming, chunk size=5,120 ms2026.06 | 3.64 | |
| Parakeet-0.6B-TDT-v22026.03 | 3.7 | |
| distil-Whisper-large-v3Size/h=22,000, Params/B=0.75, Evaluation Protocol=short-form2026.01 | 3.86 | |
| Whisper-large-v3Size/h=680,000, Params/B=1.50, Evaluation Protocol=short-form2026.01 | 3.88 | |
| FastWhisper-largeSize/h=1,634, Params/B=0.74, Evaluation Protocol=short-form2026.01 | 3.89 | |
| whisper-medium.en2026.03 | 3.9 | |
| parakeet-rnnt-1.1b2024.11 | 3.92 | |
| Whispering-LLaMa-7bparameters=7B2024.11 | 4.6 | |
| Fast Conformervariant=FT+LCA+GT2026.06 | 4.98 | |
| SpeechStew2024.11 | 5.3 | |
| whisper-large-v32026.03 | 8.1 | |
| Hikari2026.03 | 8.6 | |
| Whisper-CDModel=Large-v3-Turbo2026.03 | 10.11 | |
| Whisper BaselineModel=Large-v3-Turbo2026.03 | 12.93 | |
| Whisper-CDModel=Large-v32026.03 | 25.62 | |
| Whisper BaselineModel=Large-v32026.03 | 66.42 |