Image to Text (OCR)

Pull editable text out of photos, screenshots and scans — in 90+ languages, without uploading anything.

Languages in the image

English

Each language downloads once (2–15 MB) and is cached.

Drop files here or click to browseYou can also paste with Ctrl V

Extracted text

How to use Image to Text (OCR)

90+ languagesVietnamese with full diacritics, English, Chinese, Japanese, Korean, Thai, Arabic, Hindi, Khmer and many more.
Batch & pasteRecognize several images in one go, or paste a screenshot directly.
Private OCRRecognition runs on your device — sensitive documents never leave it.

How accurate is it?

Clean screenshots and printed documents are recognized very accurately. Handwriting, low resolution and busy backgrounds reduce accuracy.

Why is the first run slower?

Each language’s recognition data (2–15 MB) downloads the first time and is cached afterwards.

Are my images uploaded?

No. The OCR engine runs in your browser; only the language data files are downloaded.

Read a QR code from a screenshot, photo or your camera — see the link, text or Wi-Fi details inside.

Cut out exactly the part of a photo you want. Free-form or fixed ratios for Instagram, TikTok, YouTube and more — right in your browser.

Count words, characters, sentences and paragraphs as you type, with estimated reading and speaking time.

Shrink JPG, PNG and WebP files by up to 90% while keeping them sharp. Batch-compress in seconds, right in your browser.

Change the width and height of your images in pixels or percent. Batch-resize photos while keeping the aspect ratio.

Cut people out of photos automatically — get a transparent PNG or a new background in seconds.