Image to Text (OCR)

Copy Text From an Image — Free Online OCR

Need to copy text from a picture instead of retyping it? Drop any image here and this free OCR tool reads the words and hands them back as plain, selectable text you can copy and paste anywhere. It works on screenshots, phone photos of documents, scanned pages, whiteboard snaps, error dialogs, receipts, slide decks, book pages, and social posts — anything where the text is locked inside a JPG, PNG, WebP, GIF, BMP, or TIFF.

The recognition engine is Tesseract, the open-source OCR project Google maintains, compiled to WebAssembly so it runs directly on your device. Nothing is uploaded. That means a screenshot of a private chat, a payslip, or an internal dashboard is read locally and never touches a server — and because there is no upload step, results come back in a second or two rather than after a round trip.

It is built for volume, not one file at a time. Drop a folder of photos and every image is read in parallel across multiple engine instances, with per-image progress and a combined text panel you can copy in one click, download as a .txt, or export as a ZIP with one text file per image. You can also just press Ctrl+V to paste a screenshot straight from your clipboard — the fastest way to turn a screenshot into text.

How to Copy Text From an Image or Screenshot

1. 1

   Add your images

   Drag and drop one or many pictures onto the drop zone, click to browse, or press Ctrl+V (Cmd+V on Mac) to paste a screenshot directly from your clipboard.

2. 2

   Pick the language

   English is selected by default. Choose another of the 30+ supported languages — including Hindi, Marathi, Kannada, Malayalam, Tamil, Arabic, Chinese, and Japanese — if your image is not in English.

3. 3

   Let it read

   Recognition starts automatically the moment an image is added. Each image shows live progress, a word count, and a confidence score as it finishes.

4. 4

   Copy or download the text

   Copy the extracted text with one click, save it as a .txt file, download a ZIP with one file per image, or export a searchable PDF that keeps the original picture with a hidden text layer.

Why Use This Image to Text Converter?

• Batch many images at once

  Drop a whole folder and every picture is read in parallel across multiple OCR engines — not queued one at a time like most online converters.

• Paste a screenshot with Ctrl+V

  Take a screenshot, press Ctrl+V on this page, and the text is ready to copy. No saving a file first.

• 30+ languages

  Reads English plus Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Malayalam, Gujarati, Punjabi, Urdu, Arabic, Chinese, Japanese, Korean, Russian, and more — and can detect English mixed into another language.

• Nothing is uploaded

  Tesseract runs as WebAssembly inside your browser, so private screenshots, IDs, invoices, and internal documents never leave your device.

• Tuned for hard images

  Small images are automatically upscaled for a cleaner read, and optional contrast enhancement and light-on-dark inversion handle photos, dark-mode screenshots, and faded scans.

• Searchable PDF export

  Turn a scanned page into a PDF that looks identical but is fully searchable and selectable, with the recognized text layered invisibly over the image.

Turn a Screenshot Into Text in Seconds

Screenshots are the most common reason people need OCR. An error message you cannot select, a chat thread, a table inside a shared image, a code snippet someone pasted as a picture, a slide from a recorded presentation — the information is right there but not copyable. Paste it here with Ctrl+V and the words come straight back as text.

Because the engine is already warmed up before you paste, there is no loading step between dropping an image and seeing the text. Screenshots also happen to be the easiest input for OCR: crisp, high-contrast, screen-rendered type reads at very high accuracy, so a screenshot to text conversion is usually near-perfect out of the box.

Read Text From Photos and Scanned Documents

Photos are harder than screenshots — uneven lighting, shadows, camera angle, and paper texture all get in the way. If a photo comes back messy, switch on Enhance contrast, which converts the image to grayscale and stretches the tonal range so the letters separate cleanly from the background. For a dark-mode screenshot or a white-on-black sign, turn on Light text on dark to invert it first.

The Page layout setting matters too. Auto suits most documents. Single block works best for one solid paragraph, Single line for a caption or a single row, and Sparse for scattered text like UI labels, memes, diagrams, and product packaging. Trying a different layout mode is often the difference between garbled output and a clean read.

Image to Text in Hindi, Marathi, Kannada, Malayalam and More

Indic scripts are fully supported. Select Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Malayalam, Gujarati, Punjabi, or Urdu and the matching trained language model is downloaded once, then cached in your browser so every later conversion is instant and works even offline.

Real documents in these languages are frequently bilingual — Devanagari or Malayalam body text with English names, numbers, and technical terms mixed in. Tick Also detect English alongside your chosen language and the engine recognizes both scripts in a single pass, instead of mangling the English words.

Frequently Asked Questions