Runs in your browser. Your files and text never leave your device - nothing is uploaded.
Extract Text From an Image Online
The Tools Kit's free Image to Text converter uses OCR (optical character recognition) to pull English text out of a photo, screenshot or scanned page and turn it into text you can edit, copy and paste. Drop an image above, wait for the progress counter to finish, and the words from the picture appear in an editable box, so you never have to retype a receipt, a slide or a page of notes by hand.
Recognition runs in your browser with Tesseract.js, a widely used open-source OCR engine, so your image is not uploaded to a server. On first use the browser downloads the engine and the English language data from a public CDN; after that, the picture you choose is read entirely on your own device.
How to Convert an Image to Text
- 1
Add your image
Click the upload area or drag a photo, screenshot or scan onto it. JPG, PNG and WebP images work best, and one image is read at a time.
- 2
Let OCR read the text
Recognition starts automatically. A progress percentage shows while the English OCR engine reads your picture in the browser.
- 3
Check and correct the result
The extracted text appears in an editable box below a preview of your image, so you can compare the two and fix any misread characters.
- 4
Copy the text
Click Copy Text to put the result on your clipboard, ready to paste into a document, email, spreadsheet or search box.
There is no account, no watermark and no daily limit. To read another picture, simply drop the next image and the tool starts over.
How OCR Turns a Picture Into Text
To a computer, a photo of a page is just a grid of colored pixels; it has no idea that some of those pixels form the word "invoice". OCR bridges that gap. The engine first finds the regions of the image that look like lines of text, then a neural network (Tesseract's LSTM engine, which this tool uses) reads each line as a sequence of characters, using a trained English model to decide which letters and words are most likely.
The output is plain text with line breaks. Layout, fonts, bold and italic, tables and embedded pictures are not reproduced, which is why the result lands in an editable box: you can fix a misread character or join wrapped lines before you copy it. If you need the text in a different case afterwards, the case converter can switch it to sentence case or lowercase, and the word counter will tell you how long the extracted passage is.
Tips for More Accurate OCR Results
Most OCR errors come from the photo rather than the engine. A few minutes spent on the source image usually does more than anything else:
- Shoot straight-on. Hold the camera parallel to the page. Text photographed at a steep angle reads badly. If the picture is sideways or upside down, turn it upright first with the image rotator.
- Crop to the text. Backgrounds, logos and photos next to the text can be misread as stray characters. Trim the picture down to the paragraph you need with the image cropper.
- Boost the contrast. Faint print, gray-on-gray text or a shadow across the page lowers accuracy. Raising contrast and brightness in the photo editor often rescues a dim photo.
- Keep the resolution. Small, blurry or heavily compressed text is the most common cause of nonsense output. Use the original photo rather than a thumbnail, and move the camera closer to small print.
- Use a common format. JPG, PNG and WebP decode reliably. If an iPhone photo in HEIC format will not read, convert it with HEIC to JPG and try again.
Image to Text vs. PDF Text Extraction and Translation
This tool reads text that exists only as pixels. Many PDFs, by contrast, already contain a real text layer that you can select with your cursor; for those, the PDF to text tool copies the text out directly, which is faster and more accurate than OCR. This page does not accept PDF files, so for a scanned PDF, save the page you need as an image first and drop that image here.
Image to Text reads English only and keeps the text in English. If you want the English text in a photo turned into another language, such as Spanish, French, Hindi or Arabic, the photo translator runs the same English OCR and then translates the result. Browse every other converter and editor in our image tools collection.
Common uses
- Copying text out of a screenshot of a message, error dialog or web page
- Digitizing a printed letter, receipt, recipe or book page you photographed
- Pulling a quote or paragraph from a presentation slide or infographic
- Capturing the text on a poster, notice board or whiteboard after a meeting
- Getting editable text from a scanned page so you can search, edit or reuse it
Image to text terms explained
| Term | What it means here |
|---|---|
| OCR | Optical character recognition: software that finds letters in a picture and turns them into real, editable text |
| Tesseract | The open-source OCR engine this tool uses, running in your browser through Tesseract.js |
| LSTM engine | The neural-network recognition mode Tesseract uses here, which reads whole lines rather than isolated letters |
| Language data | The trained English model file the engine downloads on first use so it knows what English characters and words look like |
| Contrast | How strongly the text stands out from its background; dark text on a plain light background reads most reliably |
| Resolution | How many pixels make up each character; tiny or heavily compressed text is the most common cause of misreads |
Frequently Asked Questions
Drop a JPG, PNG or WebP image onto the upload area or click to choose one. OCR starts automatically and shows a progress percentage, then the recognized English text appears in an editable box you can correct and copy with the Copy Text button.
Yes. It is free to use with no signup and no watermark, and you can convert as many images as you like, one at a time.
No. Text recognition runs inside your browser with Tesseract.js, so the image itself never leaves your device. The first time you use the tool, your browser downloads the OCR engine and the English language data from a public CDN, and that language file is usually cached for later visits.
English only. The tool loads the English language model, so text in other alphabets such as Arabic, Chinese, Cyrillic or Devanagari will not be recognized correctly, and accented letters in languages like French or Spanish may come out wrong.
Use JPG, PNG or WebP. The upload area accepts any file the browser labels as an image, but other formats are not guaranteed to decode. If an iPhone HEIC photo fails, convert it to JPG first and try again. PDFs are not accepted.
Yes, and screenshots are usually the easiest case: the text is sharp, straight and high-contrast, so the OCR result is often nearly perfect. Crop out menus and icons around the text you want for the cleanest output.
Only neat, clearly separated block lettering has a reasonable chance. The English model is trained mainly on printed text, so joined-up or cursive handwriting usually comes out with many errors or not at all.
OCR accuracy depends on the photo. Blur, low resolution, glare, shadows, a tilted angle, busy backgrounds, decorative fonts and very small text all cause misreads. Retake the photo straight-on in even light, crop closer to the text, or rotate it upright, then run it again.
No. The tool returns plain text with line breaks. Fonts, bold and italic, columns, tables and images are not preserved, so multi-column pages may need tidying after you paste them.
No. It reads one image at a time; drop the next image to start again. For a PDF that already contains selectable text, a PDF text extractor is faster and more accurate than OCR.
Yes. Open the page in your mobile browser, tap the upload area and choose a photo from your gallery or files. Recognition runs on the phone itself, so a large photo can take a little longer than on a laptop.
Reviewed by Raja Jahangir · Last reviewed: October 7, 2026
How do you extract text from an image?
To extract text from an image, upload a JPG, PNG, or WebP and optical character recognition runs automatically, showing a progress percentage while it reads. The recognised English text appears in an editable box you can copy. OCR runs locally in the browser using Tesseract.js, so the photo is never uploaded to a server.
Why is OCR not reading my image correctly?
OCR accuracy depends on the source image. Clear, high-contrast text on a plain background reads almost perfectly, while blurry photos, skewed angles, handwriting, decorative fonts, or low resolution cause errors. Only English is recognised. Retake the photo straight-on in good light, or crop closer to the text, and the result improves substantially.