How do I use the camera feature?

On a mobile device, tap "Use Camera" to open your device's native camera directly. On desktop, where there's usually no camera pointed at paper documents, use the upload option and select a photo you've already taken.

What makes a photo scan well versus poorly?

Hold the phone steady and roughly parallel to the page, use even lighting without harsh shadows or glare, and keep the document as flat as possible. Blurry, tilted, or dimly lit photos are the single biggest cause of poor OCR results.

Can it read handwriting, like a note or a whiteboard?

Tesseract is optimized for printed and typed text. It can sometimes recognize very neat, clearly separated handwriting, but cursive or inconsistent handwriting produces unreliable results — treat handwriting recognition as a bonus, not a guarantee.

Is my photo uploaded anywhere when I use the camera?

No — the captured photo is handed straight to the in-browser OCR engine and never transmitted to a server. Everything happens locally on your device.

Which languages does it support?

The current version is tuned for English text recognition. Other Latin-script languages often produce reasonable results too, but accuracy is highest for English.

Why does the text come out jumbled when the photo looks fine to me?

A camera angle that looks fine to the human eye can still be enough skew to confuse OCR segmentation, and glare invisible in a quick glance can wash out contrast in specific areas of the frame. Retake the photo straight-on with the document as flat and evenly lit as possible.

Can I use this on a desktop computer without a camera pointed at paper?

Yes — the upload option works the same way as any file picker. Take the photo on your phone, transfer or take a screenshot, then upload it here from your computer.

What is the difference between this and the Image to Text tool?

Scan to Text is built around live camera capture for physical documents in front of you right now. Image to Text is built for images you already have saved, with no camera-specific capture flow.

Scan Document to Text — Extract Text from Phone Camera Photos

FREE

Loading…

Go Pro ↑

Upload an image

JPG, PNG, WEBP, BMP, TIFF

Your files never leave your device. All processing happens locally in your browser — nothing is uploaded to our servers.Files are never stored on our servers · No tracking of your file contentsPrivacy Policy

What is Scan to Text?

Scan to Text is built around the single most common way people actually encounter paper text that needs digitizing: they're standing in front of it, phone in hand, and they want the words out of it right now — not later, not after finding a flatbed scanner, not after saving a photo and remembering to come back to it. Tapping "Use Camera" opens your device's native camera directly from the browser, so a receipt on a counter, a letter on a desk, or a notice on a wall becomes editable text in the same motion as taking the photo. That's the meaningful difference between this tool and Image to Text on this site: Image to Text is built for images you already have saved somewhere, while Scan to Text is built for the moment of capture itself.

The one-shot, in-the-moment nature of this workflow shapes what it's realistically good at. A phone camera photographing a physical document at an angle, under whatever lighting happens to be available, is a fundamentally messier input than a flatbed scan or a clean screenshot — there's often a slight skew, uneven lighting across the frame, a bit of motion blur, or a curved page if the paper isn't held flat. Tesseract, the open-source OCR engine (originally developed at HP, later open-sourced and now maintained with backing from Google) that powers the recognition here, includes preprocessing steps specifically to compensate for some of this — deskewing a tilted capture, normalizing uneven brightness — but there's a real limit to how much correction can fix. The practical upshot: a receipt photographed flat under a ceiling light will OCR far better than the same receipt photographed at a steep angle in a dim restaurant, and knowing that in advance is more useful than being surprised by it afterward.

The use cases this tool is actually built around tend to share a "capture it before you lose it" quality: a paper receipt that needs its total and date logged into an expense report before it gets thrown away, a printed sign or menu whose text you want to paste into a translation app while you're still standing in front of it, a page from a physical book you want to quote accurately without retyping it by hand, or a handwritten note on a whiteboard you want to preserve as text before someone erases it. In every one of these, the phone camera is genuinely the fastest and often only practical way to get the source text, and OCR is what turns that photograph into something you can actually use — search, paste, translate, or archive.

As with every OCR tool on this site, the entire recognition pipeline runs locally: the photo is decoded and analyzed by Tesseract compiled to WebAssembly directly inside your browser tab, and it is never transmitted anywhere. That matters more here than almost anywhere else on the site, since a phone camera pointed at "whatever's in front of you" very often captures something you'd rather not have pass through a third-party server — a medical form, a personal letter, a handwritten note with private details.

How It Works

Tapping "Use Camera" invokes your device's native camera capture through the browser's file input API, handing the resulting photo directly to the page without any intermediate upload step. That image is then passed to Tesseract.js, the WebAssembly build of the Tesseract OCR engine, which applies preprocessing tuned for imperfect real-world photo capture — grayscale conversion, contrast normalization to compensate for uneven lighting, and skew correction for a camera angle that isn't perfectly square to the page. The corrected image is segmented into lines and words, each character shape is matched against Tesseract's trained recognition model, and the assembled text is displayed directly in an editable box on the page. Every step, from camera capture through recognition, executes locally in your browser; nothing is ever uploaded to a server.

How to Use

Upload your scanned document

Drop a scanned image (JPG, PNG, TIFF) or a scanned PDF onto the upload area.

Choose language

Select the language(s) present in the scan to optimize OCR recognition accuracy.

Run OCR

Click "Extract Text". The Tesseract.js engine processes each page image and recognizes the text characters.

Review the extracted text

The result appears in an editor. Check for recognition errors — common in low-resolution or handwritten scans.

Copy or export

Copy the text to use in Word, emails, or other tools, or download it as a .txt file.

Who Uses This Tool

🧾

Digitize Receipts on the Spot

Photograph a paper receipt right where you receive it and extract the text to log amounts into an expense report before it gets lost.

📬

Transcribe Physical Letters

Scan a printed letter the moment it arrives and extract the text for quoting, forwarding by email, or archiving digitally.

📖

Quote Book Passages Accurately

Photograph a passage from a physical book while you're reading it and extract the exact wording for research or academic citation.

🪧

Capture Signs and Notices

Point your camera at a public sign, restaurant menu, or notice board to extract the text instantly rather than retyping it by hand later.

📄

Digitize Completed Paper Forms

Photograph a filled-in paper form on the spot to extract the entered data and copy it directly into a digital system or spreadsheet.

🗒️

Preserve Whiteboard Notes

Scan handwritten meeting notes from a whiteboard before they get erased, turning them into searchable, storable text.

Why Choose This Tool

Camera Capture Built In

Tapping "Use Camera" opens your phone's camera directly from the browser, turning capture and recognition into a single motion — no separate scanning app needed.

Tuned for Imperfect Real-World Photos

Preprocessing compensates for the skew and uneven lighting that come with photographing a physical document by hand, rather than assuming a perfectly flat scan.

No Upload, Ever

All OCR processing runs locally in your browser via WebAssembly — your photos and the extracted text never reach any server.

Multiple Image Formats Accepted

Accepts JPG, PNG, WEBP, TIFF, and HEIC images, covering photos taken by any modern smartphone or camera.

Copy or Download Output

Copy the extracted text to your clipboard in one click, or download it as a plain .txt file for later use.

Free to Start — Sign Up for More

Use free with no account for files up to 40 MB. Sign up free to raise your limit to 80 MB, or upgrade to Pro for 250 MB per file.

Key Benefits

Opens your phone's camera directly from the browser for instant capture-to-text

No upload — OCR runs entirely locally using Tesseract compiled to WebAssembly

Built-in preprocessing compensates for camera skew and uneven real-world lighting

Works well on receipts, letters, book pages, signs, and handwritten whiteboard notes

Copy the result instantly or download it as a .txt file

Also accepts existing photos via upload on desktop, not just live camera capture

Supports multiple image formats: JPG, PNG, WEBP, HEIC

Frequently Asked Questions

Free Online Document Tools for Everyone

iLoveDocsTools is used by students, professionals, and businesses across the United States, Canada, United Kingdom, Australia, India, and more than 100 other countries. All tools are browser-based and process files privately — your files never leave your device. Free to use, with Pro plans for unlimited access.

Scan to Text

What is Scan to Text?

How It Works

How to Use

Who Uses This Tool

Why Choose This Tool

Key Benefits

Frequently Asked Questions

Related Tools

Free Online Document Tools for Everyone

No Upload Required

Free to Use — Pro Plans Available

Works on Any Device