best model by language
Benchmarked on the FLEURS test split at Q8_0. Error is WER — lower is better. Speed is × realtime on Handy target hardware.
accuracy vs speed
Lithuanian · 14 models
AMD Ryzen 4750U GPU (Vulkan)
pareto frontier (no model is both faster and more accurate) other models
highlighted rows are recommended picks or on pareto frontier
| | | | | | |
| 1 | canary-1b-v2!overall pickaccurate pickfast pick | 13.45 | 14.1× | 8.3× | 1.07 GB | CC-BY-4.0 |
| 2 | parakeet-ultra | 17.06 | 23.0× | 13.3× | 0.69 GB | CC-BY-4.0 |
| 3 | orukeet | 17.72 | 25.1× | 13.6× | 0.69 GB | CC-BY-SA-4.0 |
| 4 | parakeet-tdt-0.6b-v3 | 22.20 | 25.1× | 13.7× | 0.69 GB | CC-BY-4.0 |
| 5 | parakeet-primeline | 23.08 | 25.0× | 13.6× | 0.69 GB | CC-BY-4.0 |
| 6 | whisper-large-v3 | 24.55 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 7 | whisper-large-v3-turbo | 25.11 | 3.1× | 1.4× | 0.83 GB | Apache-2.0 |
| 8 | whisper-large-v2 | 30.37 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 9 | whisper-large | 37.58 | 2.0× | 1.0× | 1.55 GB | Apache-2.0 |
| 10 | whisper-medium | 43.34 | 4.1× | 1.9× | 0.77 GB | Apache-2.0 |
| 11 | whisper-small | 70.21 | 11.7× | 5.9× | 0.25 GB | Apache-2.0 |
| 12 | Fun-ASR-MLT-Nano-2512! | 78.42 | 8.8× | 5.5× | 0.83 GB | FunASR Model Open Source License Agreement v1.1 |
| 13 | whisper-base | 91.78 | 29.0× | 19.3× | 0.08 GB | Apache-2.0 |
| 14 | whisper-tiny | 105.00 | 43.1× | 39.8× | 0.04 GB | Apache-2.0 |