Image to Text
Reads printed text out of an image and hands you plain text you can copy. The recognition engine is downloaded into your browser once, then every image is read on your own machine.
- No upload
- No account
- No watermark
- No file limit
Your files
Nothing chosen yet.
Where your file goes
Nowhere. This page downloads a small program into your browser, and that program reads the file straight from your disk. The program comes from this site like any image or stylesheet — but your file only ever travels from your disk into your own browser tab.
Why this one
What you get that the big converters cannot offer.
The image never leaves
Screenshots often hold things you would rather not hand over — an invoice, a chat, an ID card. There is no server here to hand them to.
No account, no page count
Run it on one image or two hundred. Nothing is metered, because nothing costs us anything to run.
Works offline after the first run
Once the engine is cached, the page keeps working with your network disconnected. You can test that yourself.
Plain text out
You get the text itself, not a PDF wrapped around a picture. Copy it or save it as .txt.
How to use it
Choose one or more images.
Wait while the recognition engine loads. This happens once and is a few megabytes.
Copy the text, or save it as a .txt file.
When people use it
Copying text from a screenshot
Someone sends a screenshot instead of the text. This gets the words back out so you can paste them.
Typing up a printed page
Photograph the page, run it through, then fix the few words it gets wrong. Faster than retyping.
Pulling numbers off a receipt
Useful for expenses — and exactly the sort of document you would not want sitting on a stranger's server.
Questions
Is my image uploaded?
No. There is no server behind this page. Recognition runs on your own computer, which is also why the first run takes a moment to start.
Why is the first run slow?
The recognition engine has to download once, about four megabytes. After that your browser caches it and later images are much quicker.
Which languages can it read?
English, Simplified Chinese and Traditional Chinese — pick one above the drop zone. Accuracy depends on how sharp the text is; a clean screenshot reads far better than a photo taken at an angle.
Can it read handwriting?
Not reliably. The engine is built for printed text.
Is there a limit on how many images?
No. The only limit is your own memory.
Other tools
Compress Image
Make photos smaller without the upload
Compress PDF
Shrink a PDF without uploading it
Compress Video
Make a video file smaller
Crop Video
Drag a box over the picture and keep only that part
Video to GIF
Turn part of a video into an animated GIF
Merge PDF
Join several PDFs into one
Split PDF
Pull pages out of a PDF
PDF to JPG
Turn every page of a PDF into an image
PDF to Word
Turn a PDF into an editable .docx
Audio Cutter
Trim an audio file
Video Cutter
Trim a video without re-encoding it
Crop Image
Drag a box over it, then rotate if it needs it
Combine Images
Put several pictures into one file
HEIC to JPG
Convert iPhone photos to JPG
JPG to PDF
Turn JPG and PNG images into one PDF
Video to MP3
Pull the audio out of a video
Word to PDF
Turn a .docx into a PDF, or one PDF per page
Markdown to Word
Turn a .md file into a .docx
Markdown to PowerPoint
Turn a Markdown outline into slides
Split Excel Sheets
One sheet, one file
Split Excel Rows
One row, one file