Translate Text Inside Scans and Images
Lingivio uses OCR to read text from scanned PDF pages and supported image files, translates it, and places the translated content into the output.
What Stays Intact
- The supported page or image layout
- Surrounding visual content
- Numbers, codes, URLs, and email addresses
- The original supported output format
- A clear credit cost before processing begins
What Survives When You Translate PNG, JPEG and WebP images
- Text is read by OCR at two scales and the results are merged, because recall is not monotonic in resolution and each pass finds lines the other drops.
- The background under replaced text is repaired texture-aware, so a photograph or a gradient behind a caption is not flattened into a grey rectangle.
- Solid graphics under a label — an arrow, a rule, a filled shape — are rebuilt rather than erased along with the text.
- Where OCR cannot read a region, the original pixels stay: an untranslated line, never an invented one.
How It Works
- 1.
Upload the Scan or Image
Add a scanned PDF, JPG, PNG, or WebP file. Lingivio checks the file and identifies the text that requires OCR.
- 2.
Choose the Target Language
Select from the languages supported for PDF and image output and review the exact cost.
- 3.
Download the Translated Result
Receive the translated file with its supported surrounding visual content preserved.
Plans and Credit Cost
Scanned PDF and image translation is included with Business and Enterprise plans. Each scanned PDF page uses 2 credits. JPG, PNG, and WebP files use 2 credits per image.
Compare PlansSee It on a Real Document
The before and after below is a real PNG, JPEG and WebP images file this engine translated (English to Spanish). The examples page has the files themselves.
What This Does Not Do
- Scanned pages and images are best-effort: on dense or degraded scans a share of lines is never extracted at all. Those lines stay in the source language on the page rather than being guessed at, and the page is charged in full whatever OCR reads off it. We do not promise a scan comes back 100% translated. Born-digital pages carry no such caveat.
- Handwriting-style Nastaliq — Urdu in particular — is not read reliably by any OCR we have tested. Those files fail with an explicit error instead of returning something that merely looks translated.
- Anything the verify stage cannot confirm falls back to the source text, visible and untranslated. A file that looks finished but is not is the one outcome we will not ship.
Questions
- Which image formats can I translate?
- Lingivio supports JPG, PNG, and WebP image files, as well as scanned PDF pages.
- Why do scans and images use more credits?
- Their text must first be read using OCR before it can be translated and placed into the output.
- Which languages are available?
- Scanned PDF and image output currently supports 20 target languages. Chinese, Japanese, Korean, Arabic, and Hindi are not currently available for these output formats.
- Is OCR always perfect?
- OCR quality depends on the source image. Clear, high-resolution text produces the strongest result, and important content should be reviewed before use.
- What happens to my file afterwards?
- Uploaded and generated files are automatically deleted within 24 hours, and are never used to train a model.
Translate Text Inside Scans and Images
Upload your file to see the supported languages and exact credit cost before translation begins.
Free credits every month · no card required