Free · no account · browser only
Read text from images without uploading them
Read text from photos and screenshots in your browser. Pick a language, then edit, copy, or download .txt. A batch adds a ZIP. First use downloads the reading engine and that language's data once. Your images are never uploaded.
Your images are never uploaded.
- 1515 languagesEnglish by default. Deutsch and Español one tap away.
- 12Up to 12 imagesOne .txt each, or one ZIP for the batch.
- 1×First-use downloadEngine + language data, once.
Image to Text
Drop, open, or paste images. Text appears as soon as it is read.
Images stay on your device · not cloud OCR
Preview
First use downloads the reading engine and English data once (about 7 MB) from jsDelivr. Switching language may download that language once. Your images are never uploaded.
Add an image to get a .txt file.
/image-to-text
How it works
Five steps from drop or paste to .txt. Pick a language first. The engine and each language's data download once. Your images are never uploaded.
- 1
Add images
Drop or paste JPG, PNG, WebP, or HEIC. Add up to 12 images for a batch.
- 2
Pick a language
Starts on English. Tap Deutsch or Español, or choose from 15 languages. Pick the language in the image before reading.
- 3
Wait for the text
Reading runs in your browser. First use downloads the OCR engine once. The first time you pick a language, its data downloads once from jsDelivr.
- 4
Edit or copy
Recognized text stays editable. Fix a line, or tap Copy text.
- 5
Download .txt
Download sits under Preview. A batch adds ZIP with one .txt per image.
Frequently Asked Questions
Honest answers about privacy, the first download, languages, ZIP, and accuracy.
Are my images uploaded for OCR?
No. Text recognition runs in your browser with Tesseract, an open-source OCR engine. Your images are never uploaded to NixRey or anyone else. The .txt and ZIP files are created on this device.
Why does the page download files the first time?
To read text, your browser needs the OCR engine and data for the language you pick. On first use it downloads the engine once from jsDelivr, a public CDN (about 4 MB). Each language then downloads its own data once, the first time you pick it: a few MB per language, about 3 MB for English. That download is software and language data only. Your images are not sent anywhere. After that, the browser can reuse its saved copy.
Which languages can it read?
15 languages: English, Deutsch, Español, Français, Italiano, Português, Nederlands, Polski, Türkçe, Русский, العربية, हिन्दी, 日本語, 한국어, 简体中文. Pick the language in your image before reading. If you pick the wrong one, text comes out wrong or empty. Each language's data downloads once, the first time you use it.
Can I fix the text before I download it?
Yes. The recognized text stays editable. Fix a line, copy it, or download it as .txt.
Can I convert many images at once?
Yes. Add up to 12 images. Each one gets its own text. ZIP under Preview packs one .txt per image. It is not an image archive.
Which image formats work?
JPG, PNG, WebP, and HEIC photos and screenshots. HEIC works when your browser can read it. If it cannot, convert it with HEIC to JPG first. GIF and AVIF also work when the browser can decode them (GIF uses the first frame).
How accurate is it?
Clear, printed text in the language you picked works best: screenshots, documents, signs, and menus shot straight on in good light. Quality is not the same across every language or script. Blur, tilt, shadows, small or decorative fonts, tables, and multi-column pages cost accuracy. Handwriting is not a strength. Proofread names, numbers, and codes before you reuse the text.
Is this as good as Google Lens?
Not always. Google Lens reads photos on Google's servers and handles more languages and handwriting. This tool reads printed text in 15 languages on your device, so your image stays private. For a private document, that trade is often worth it.
Is this cloud OCR or an AI service?
No. There is no cloud OCR and no server reading your image. Recognition runs locally in your browser.
Is it free? Do I need an account?
Free, with no sign-up and no watermark.
Can it read a PDF?
Not on this page. If your PDF already has selectable text, use PDF to Text. Scanned PDFs need OCR on each page, which this page does not do.
My photo shows a GPS warning. What does that mean?
Some photos carry location data. Reading the text does not change your photo. If you plan to share the original, use Remove EXIF Data first.