Persian translation and speech-to-text models fa
1 model into Persian, 3 from Persian, 14 speech-to-text.
Which open models work for Persian?
Windy Translate lists 1 open translation model into Persian and 3 from Persian. All of them can be downloaded from Hugging Face and run on your own hardware. By licence: 4 CC-BY-4.0 (terms and attribution text: licensing). None of them has a screening score yet, so test them on your own text before you rely on them.
With English. English to Persian: translate-windy-hplt-en-fa (not yet scored). Persian to English: translate-windy-hplt-fa-en (not yet scored). To translate between Persian and a language with no direct model, you can go through English with these two.
Speech-to-text. 2 speech-to-text models made for Persian: listen-windy-lingua-fa and listen-windy-lingua-fa-ct2. Also, Whisper's tokenizer lists Persian as one of its languages, so all 12 multilingual Whisper-based models here can transcribe it: see the speech models. Accuracy differs a lot between languages and model sizes, and this site has not measured speech accuracy.
Each model page has a short snippet to download and run the model yourself, its licence and attribution text, and its score where there is one. Pick the best scored model for your direction, then try it on a few sentences of your own text: a screening score is a starting point, not a guarantee.
Into Persian
| Model | chrF++ (FLORES-200 dev) | Licence |
|---|---|---|
| English to Persian translate-windy-hplt-en-fa | not in FLORES-200 | CC-BY-4.0 |
From Persian
| Model | chrF++ (FLORES-200 dev) | Licence |
|---|---|---|
| Persian to North Germanic languages translate-tc-big-fa-gmq | not in FLORES-200 | CC-BY-4.0 |
| Persian to Italic languages translate-tc-big-fa-itc | not in FLORES-200 | CC-BY-4.0 |
| Persian to English translate-windy-hplt-fa-en | not yet scored | CC-BY-4.0 |
Persian speech-to-text
| Model | Formats | Licence |
|---|---|---|
| listen-windy-lingua-fa | safetensors | Apache-2.0 |
| listen-windy-lingua-fa-ct2 | ct2-int8 | Apache-2.0 |
Scores: chrF++ on FLORES-200 dev, 48 sentences, measured 2026-08-04. Method.