Automatic Speech Recognition on Adversarial
17.5WERparakeet-rnnt-1.1b
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| parakeet-rnnt-1.1bparameters=1.1B, architecture=RNNT2025.02 | 17.5 | 17.2 | 29.8 | 30.6 | — | |
| parakeet-tdt-1.1bparameters=1.1B, architecture=TDT2025.02 | 19.4 | 19.1 | 32.2 | 30.2 | — | |
| canary-1bparameters=1B2025.02 | 27.9 | 27.5 | 46.3 | 41.6 | — | |
| Qwen2-Audio-7BArchitecture=Qwen2, Parameters=7B2025.02 | 31.7 | — | — | — | 46.7 | |
| whisper-large-v3Model Size=Large, Version=v32025.02 | 33.3 | — | — | — | 49.8 | |
| whisper-mediumModel Size=Medium2025.02 | 42.9 | — | — | — | 63.1 | |
| wav2vec2-large-xlsr-53-englishModel Size=Large, Architecture=wav2vec2-xlsr2025.02 | 100.1 | — | — | — | 96.5 | |
| hubert-large-ls960-ftModel Size=Large, Finetuning=ls9602025.02 | 109.8 | — | — | — | 100 |