Offline translator
Translate text between English and Spanish with local browser models.
Offline translator
Sådan bruger du Offline translator
Offline translator translates text between English and Spanish with Marian (opus-mt) models that download on first use and then run in your browser. Set the source and target language codes to en and es, in either direction, and translate up to 30,000 characters per run. The text is processed in 500-character pieces, and the result downloads as translated.txt.
- Paste the source into the input editor, or load the built-in example.
- Set source language code, target language code.
- Run offline translator, review the output, then use the available copy or download controls.
Værktøjets funktioner
English ↔ Spanish translation uses browser Marian models, downloaded on first run. Document translation exports extracted text in Markdown and does not rebuild page layout.
- Source language code
- Set this value for your task.
- Target language code
- Set this value for your task.
Grænser og behandling
Image inputs: up to 32 megapixels, batches up to 10 images. Audio/video uploads: 50 MB; transcription: 120 seconds. Edited videos: 20 seconds, 4 fps and 640 pixels wide. Translation: English/Spanish, 30,000 characters; subtitles: 200 cues. Neural speech: 1,000 English characters. Model downloads and memory requirements vary by device.
Eksempel på input
A useful tool makes everyday work easier. Review the results before sharing.
Ofte stillede spørgsmål
Which languages does the offline translator support?
Only English to Spanish and Spanish to English. Enter en and es as the source and target codes, in either order; any other pair returns an error explaining that browser translation currently supports English and Spanish.
Why did the translation change my line breaks?
Text is cut into 500-character pieces, each piece is translated separately, and the results are joined with line breaks. Pieces can end mid-sentence, so check the joins and restore paragraph breaks where needed.
Does the offline translator need an internet connection?
The first run downloads the translation model from Hugging Face. After that the model may remain in the browser cache and translation runs locally; if the cache is cleared, the model is downloaded again on the next run.
Where is my input processed?
Source files and text stay in this browser. The first run downloads public models from Hugging Face and local runtime assets from the site. Models may remain in the browser cache. No source upload to Vercel, Supabase or a conversion service is required.
What are the input limits?
Image inputs: up to 32 megapixels, batches up to 10 images. Audio/video uploads: 50 MB; transcription: 120 seconds. Edited videos: 20 seconds, 4 fps and 640 pixels wide. Translation: English/Spanish, 30,000 characters; subtitles: 200 cues. Neural speech: 1,000 English characters. Model downloads and memory requirements vary by device.