Free Tools
/
Image to Text Converter
Image to Text Converter
Add a screenshot, a scan or a photo of printed text and get the text back to copy or save. The image is read in your browser and is never uploaded.
Get the text out of a picture without uploading it
Best for
Students, office workers and anyone with a screenshot, a scanned page or a photo of printed text that they want to copy, search or edit.
What it does
Add one or more images, choose the language, and press Convert. Then copy the text, or save it as a TXT or Word file.
Before you rely on it
It works best on clear printed text. Handwriting, angled or shadowed photos, low-resolution images and tables are often read badly, and the layout of columns and tables is not kept, so check the text before you rely on it.
Count the words in the text
Turn a short audio clip into text
Add your images
Up to 10 images as PNG, JPG, WebP, BMP or GIF. Choose files, drop them here, or paste a screenshot with Ctrl+V.
Drop images here, or click to choose
Your images are read in this browser and are never uploaded.
Language of the text
Convert to text
How to convert an image to text
Add your images by choosing files, dragging them onto the box, or pasting a screenshot with Ctrl+V (Cmd+V on a Mac). Choose the language the text is written in, then press Convert. The first time you convert, your browser downloads the reading engine from this site, which takes a few seconds on a normal connection. After that it starts faster.
When it finishes, the text appears in a box you can edit. Copy it, or save it as a plain text file or a Word document.
- You can add up to 10 images at once, each up to 15 MB, as PNG, JPG, WebP, BMP or GIF files.
- Phone photos in the HEIC format cannot be read. Export or save the photo as JPG or PNG first.
- With more than one image, the text from each is kept in order under the image's file name.
What it reads well and what it does not
The tool recognises printed letters. How well it does depends mostly on the image, not on the settings.
| Kind of image | What to expect |
| Screenshot of a web page, document or chat | Usually reads well |
| Dark-mode screenshot, or light text on a coloured background | Turned round automatically. Usually reads well |
| Sharp scan or straight-on photo of a printed page | Usually reads well |
| Photo taken at an angle, in shadow or with glare | Straightened and evened out automatically. Heavy glare, blur or a sharply slanted view can still cause errors |
| Low-resolution or heavily compressed image | Small text is often misread |
| Handwriting | Not supported. Results are usually poor |
| Tables, forms and columns | The words are read, but the layout is not kept |
| Printed text in Hindi, Kannada, Tamil, Telugu, Malayalam, Bengali or Gujarati | Works on clear text when the language is chosen, with more mistakes than English |
| A thermal-printed bill or receipt photographed on a phone (an electricity bill, a shop slip) | The hardest case: small, faint print. Expect mistakes in words and in digits, and check every figure against the paper |
What the tool does to get a clean reading
Before it reads a picture, the tool prepares the picture on your device. It enlarges small text, turns light text on a dark background round so that a dark-mode screenshot or a chat reads as dark on light, evens out shadows, and straightens a photo of a page that was taken at an angle.
If the first reading looks poor, it tries again in other ways, such as reading the picture as one block of text or turning it, and keeps whichever reading is clearest. If an English reading comes out as symbols, it also tries Hindi and English. A picture that reads cleanly is read once; a difficult one can take up to half a minute.
If nothing reads clearly, the tool says so instead of showing rows of symbols as if they were the text. You can still ask to see what it found.
Tips for a better result
- Crop the image to the text you need, so the engine is not distracted by pictures and borders.
- Use the highest resolution you have. A sharp screenshot beats a blurry photo of the same page.
- Hold the camera straight above the page and avoid shadows and glare.
- Choose the language of the text. Bills and forms in Kannada, Tamil, Telugu, Malayalam, Bengali, Gujarati or Hindi nearly always carry English too (numbers, labels, names), so pick that language together with English.
- Read the result once. Look closely at numbers, names and punctuation, which are the usual places for mistakes.
Your images stay on your device
The image is read by a program that runs inside your browser, so it is not sent to Libraryminds or to anyone else. The reading engine and the language data are loaded from this site, not from another company's server. If you close the page, the text is gone unless you copied or saved it.
Your browser may keep the language data for later visits so it does not download it again. It stores only the language files, not your images or your text.
The reading engine is Tesseract, open-source software released under the Apache License 2.0, run in the browser through Tesseract.js. Its licence files are published at /ocr-engine/licenses/NOTICE.txt.
What it does not do
- It does not read handwriting.
- It does not read PDF files. Save a page of a PDF as an image first.
- It does not translate the text or fix mistakes in it.
- It does not keep the layout, columns, tables or fonts of the original.
Frequently asked questions
How do I convert an image to text?
Add the image, choose the language and press Convert. The text appears in a box where you can edit it, copy it or save it as a TXT or Word file.
Are my images uploaded?
No. The image is read by a program that runs in your browser. The reading engine and language data are loaded from this site, and your image is never sent anywhere.
How do I copy text from a picture on my phone?
Open this page in your phone's browser, choose the picture from your gallery, and press Convert. Then use the Copy button. A photo saved in the HEIC format has to be exported as JPG or PNG first.
Why is some of the text wrong?
The tool recognises printed letters and can misread blurry, small, angled or shadowed text, as well as numbers and punctuation. Use a sharper, straighter image, choose the right language and read the result once before you use it.
Does it work with Hindi, Kannada, Tamil or other Indian languages?
Yes, for clear printed text in Hindi, Kannada, Tamil, Telugu, Malayalam, Bengali and Gujarati. Choose the language together with English, because bills and forms mix the two. Expect more mistakes than with English, and more again for small or faint print. If you leave English chosen and the picture is in Hindi, the tool notices when the English reading comes out as symbols and reads it again as Hindi and English. For the other languages it offers a button to read the picture again in that language.
Why did it say the image could not be read reliably?
The reading came out as mostly symbols, not words. The usual reasons are handwriting or a decorative font, a blurry or very small photo, glare or a shadow over the text, or a language other than the one chosen. Retake the photo straight-on in good light, crop it to the text, and choose the language (the tool offers buttons to read it again as Kannada, Tamil, Telugu, Malayalam, Bengali, Gujarati or Hindi), then convert again.
Can it read handwriting?
No. It is built for printed text, and results on handwriting are usually poor.
Can I save the text as a Word file?
Yes. After converting, choose Download Word to get a .docx file with one paragraph for each line, or Download TXT for plain text.
How many images can I add, and how large can they be?
Up to 10 images at a time, each up to 15 MB, as PNG, JPG, WebP, BMP or GIF files.
Need the Content, Not the Runtime?
Libraryminds offers transcription, AI summaries, and deep video search. Turn your video content into a searchable knowledge base today.
Try Libraryminds Free
Explore Features
Love this free tool?
Need the answer, not the whole recording? Search a transcript
Try Libraryminds Free