ESEN
Image to Text (OCR)

Convert an image to text in seconds: copy from photos, invoices, and screenshots with no limits or signup. Local AI in your browser.

Drop an image or press Ctrl+V
orbrowse your files
JPG · PNG · WEBP · SCREENSHOTSImages up to 50 MB
Or try an example:
ON-DEVICE AI · NO SIGNUP

Extract text from image online with OCR, free and unlimited

Convert an image to text in seconds: paste a screenshot with Ctrl+V or drag in a photo, receipt, or scanned document to edit, copy, or download as TXT or JSON. Your image never leaves your device.

WORKFLOW

How to convert an image to text in 3 steps

  1. 01

    Upload the image or paste it with Ctrl+V

    Drag in a photo or document, or paste a screenshot with Ctrl+V (Cmd+V on Mac). Pick a size for your image: Fast (~6 MB), Balanced (~31 MB), or High accuracy (~139 MB).

  2. 02

    Review the detected text in seconds

    Every recognized line is outlined on the image. Click any box to copy just that line.

  3. 03

    Edit, copy, or download

    Edit the text in the side panel, copy it in one click, or download it as TXT. If you need structured output, the JSON includes the coordinates and confidence of each line.

USE CASES

What it is used for

Invoices, tickets, and receipts

Copy amounts, dates, and line items into a spreadsheet without typing them by hand.

Screenshots and slides

Paste a screenshot of a video, a video call, or a locked PDF and get editable text back.

Signs, printed notes, and phone photos

Works with camera photos as long as the text is readable and not too skewed.

Structured data for developers

Download the result as JSON with coordinates (xmin, ymin, xmax, ymax) and the confidence of each line.

FAQ

Frequently asked questions about extracting text from images

How do I extract text from an image or screenshot? +
Upload your image or paste it directly with Ctrl + V (Cmd + V on Mac) anywhere on this page, wait a few seconds, and copy the recognized text. It works with photos, screenshots, and scanned documents without having to save the file first or install any software.
Is it free? Are there usage limits or signups required? +
It is 100% free with no account required: there are no daily limits or subscriptions, so you can convert as many images to text as you need. The only technical limits are a 50 MB maximum file size and your device’s available memory.
Are my images or documents uploaded to a server? +
No. Recognition runs entirely in your browser and your image never leaves your device, making it safe for invoices, contracts, or private documents. On your first visit, only the model and execution engine are downloaded to your browser cache.
What languages does the OCR support? +
It recognizes English, Spanish, and other Latin-alphabet languages, accurately handling accents, special characters, and Spanish punctuation. It is not currently designed for non-Latin scripts such as Chinese, Arabic, or Cyrillic.
Can I extract text from a scanned PDF? +
Not directly, since the tool accepts image files (up to 50 MB) rather than PDF documents. Simply take a quick screenshot of the page or convert it to an image first, then upload or paste it here to pull the text.
Which model should I choose: Fast, Balanced, or High accuracy? +
Text recognition is powered by PaddleOCR (PP-OCRv6). Balanced (~31 MB) handles most screenshots and printed pages with great speed and accuracy. Fast (~6 MB) downloads quickly and saves mobile data, but may miss finer details. High accuracy (~139 MB) is ideal for small fonts, wrinkled receipts, or low-light captures.
Does it work offline and across all browsers? +
It works on any modern desktop or mobile browser. It leverages WebGPU for hardware acceleration to finish in seconds when available, automatically falling back to WASM (CPU) otherwise. After the initial download, the model remains in your browser cache and generally works offline.
What kinds of text struggle, and how can I improve accuracy? +
Handwriting, heavily angled text, motion blur, and low contrast can cause recognition errors—always double-check numbers and proper names. For best results, use a sharp image, crop down to just the text section, and try the High accuracy model.
More

Other tools

Size

Compress Image

Optimize and reduce photo file size without quality loss.

Open tool
Framing

Crop Image

Crop areas of your image to refine composition.

Open tool
Privacy

Remove Metadata

Delete EXIF and GPS camera data to protect privacy.

Open tool
Background

Remove Background

Remove the background from a photo and download a transparent PNG.

Open tool