Image Tools
Read the text out of an image, without uploading it
Add a picture and press the button. Tesseract runs inside this page rather than on somebody else’s server, which means a payslip, a passport or a contract can go through it without being sent anywhere.
Add a photograph, a screenshot or a scan and the text is read out of it here.
Recognised in your browser. Your image is not uploaded.
- Runs in your browser
- No account required
- Free to use
- Up to 50 files
How it works
The recogniser is downloaded, not the image uploaded
Every other free OCR page posts your picture to an API. This one does the opposite: it fetches about four and a half megabytes of WebAssembly and English language data from this site, once, and then reads the image locally. Your browser caches it, so the second image is immediate.
Nothing is fetched from anywhere else
The library that does this normally pulls its worker, its engine and its language data from a public CDN. All three are served from this domain instead, because a single off-origin request would make the promise on this page untrue — and there is an automated check that fails the build if one appears.
Telling it what it is looking at helps a lot
By default it assumes a page of prose and goes looking for columns and paragraphs. On a sign, a serial number or a receipt that assumption is wrong and costs accuracy. Saying "a single line" or "a column of text" often matters more than anything about the image itself.
What it does
A confidence figure you can act on
Scored and then translated into whether it is worth reading through, checking closely, or retyping. The raw number looks like a percentage of correct characters and is not one.
Line joining, off by default
Recognition breaks a line wherever the picture does, so prose returns as a narrow column. Joining is one click — and stays off by default because on a receipt or a table the breaks are the structure.
Copy or save
Straight to the clipboard, or out as a text file.
The image never leaves your device
No account and no upload. The recogniser comes to the image rather than the image going to a server, which for a document with somebody’s name and address on it is the only arrangement worth having.
Practical advice
Getting a better result
- Crop to the text before reading it. Everything else in the frame is something for the recogniser to get wrong.
- Straighten a photographed page first — the rotator on this site will do it. Skew costs more accuracy than resolution does.
- Higher contrast beats higher resolution. A sharp black-on-white scan at 1000px reads better than a grey photograph at 4000px.
- Set the layout to match the picture. "A single line" on a sign, "a column of text" on a receipt; the default is for a page of prose.
- Handwriting will not work. Tesseract is trained on printed type, and no setting here changes that.
- English only at the moment. Another language needs its own data file, which is another few megabytes for everyone who visits the page.
Questions
Things people ask
Is my image uploaded?
No. The recogniser is downloaded to your browser and the image is read there — the traffic goes the other way round from every other free OCR page. You can confirm it in your browser’s developer tools: the engine and the language data come down, and nothing of your image goes back up.
Why is it slow the first time?
Because it is fetching the engine and the English language data — about four and a half megabytes. That happens once and is then cached; afterwards recognition takes a few seconds and no download at all.
How accurate is it?
On a clean, straight, printed page, very. On a photograph taken at an angle in poor light, poor. The confidence figure tells you which of those you have, and cropping and straightening the image is usually the fastest way to improve it.
Can it read handwriting?
No. Tesseract is trained on printed text. Handwriting recognition is a different problem and needs a different model.
Can it read a PDF?
Not directly. Convert the pages to images first — the PDF to PNG tool on this site does that — and read them one at a time. If the PDF already has a text layer, copying from it will be perfect and OCR would only make it worse.
What about other languages?
English only for now. Each language is a separate data file of a few megabytes, and adding them all would be a cost paid by everyone who opens the page.
Related tools
Image Converter
Convert between JPG, PNG, WebP and AVIF - up to 50 images at once.
Runs entirely in your browser
Open toolImage Compressor
Shrink image file size with a quality slider and see the saving.
Runs entirely in your browser
Open toolImage Optimizer
Analyse an image for web suitability, then optimise what it finds.
Runs entirely in your browser
Open toolImage Resizer
Set an exact width or height, or scale by a percentage.
Runs entirely in your browser
Open toolWho builds these
Free tools by day. Websites that earn their keep by trade.
KuConvert is built by KUDUY, a web design and development agency. From custom WordPress builds to SEO and e-commerce, everything they do is engineered to bring you more qualified traffic and turn it into revenue.
Web Design
Custom, conversion-focused web design that captivates visitors and turns them into paying customers. Mobile-first, fast, and built to rank.
Web Development
Custom WordPress websites built with clean code, blazing-fast performance (Core Web Vitals optimised), and enterprise-grade security.
E-Commerce
Online stores that convert visitors to customers with seamless shopping experiences.
SEO
Technical SEO, keyword research, on-page optimisation and local SEO strategies that drive qualified organic traffic.
Digital Marketing
Data-driven campaigns that deliver measurable ROI across every channel.
Branding
Memorable brand identities that stand out in any market and build lasting trust.
Website Help
One-on-one expert support for bug fixes, updates and ongoing maintenance — without a long-term retainer.
Fixed scope, no surprise invoices
Most sites ship in four to eight weeks with the scope and the price agreed up front.
SEO from day one
Search is designed into the build — structure, schema and Core Web Vitals — not retrofitted.
A team, not a ticket queue
A project manager, a designer and a developer you can actually reach.
Need your whole site to load this fast?
Images are usually the heaviest thing on a page and the easiest win. KUDUY works on site speed, image delivery and Core Web Vitals for teams who would rather not do it by hand.
This tool runs entirely in your browser. Your files are never uploaded.