1
Background Remover
Cut the subject out of a photo and save it as a transparent PNG, processed entirely on your own device.
Image to Text (OCR)
Turn screenshots, scanned pages and photos of printed text into editable text without sending the image anywhere.
Drop, paste or click to choose an image
Ready. The model downloads the first time you run it (~10 MB).
Optical character recognition (OCR) reads the letters in a picture and returns them as plain text you can copy, search and edit. It is useful when you have a screenshot, a receipt, a scanned letter or a slide and would rather not retype it.
Everything happens on your own computer. The image is read by code running in this browser tab and is never uploaded, so it suits documents you would not hand to an online converter.
How to use it
- Click the file picker or drag an image (PNG, JPG or WebP) onto the drop area.
- Wait for the recognition engine to load. The first time, the browser downloads about 10 MB of English language data, which is cached for later visits.
- Watch the progress bar while the page is analysed; a clear screenshot usually takes a few seconds.
- Review the extracted text, correct any misread characters, then copy it to your clipboard.
How it works
The tool uses Tesseract.js, a WebAssembly port of the open-source Tesseract OCR engine. It first cleans up the image and finds the lines and words on the page, then a trained recognition model compares the shapes of the characters against its English language data and picks the most likely letters. Because the engine is compiled to WebAssembly, it runs at close to native speed inside the browser without any plug-in.
Good for
- Copying text out of a screenshot of a chat, web page or software error dialog.
- Digitising a printed letter, invoice or receipt so you can search or archive it.
- Pulling quotes from a photographed book page or presentation slide for your notes.
- Making text inside an image available to paste into a translator or a screen reader.
Limitations
- Only English language data is loaded, so other languages and non-Latin scripts will be recognised poorly or not at all.
- Handwriting, stylised fonts, low-resolution photos, strong shadows and skewed pages reduce accuracy noticeably; always proofread the output.
- Tables and multi-column pages may come out in the wrong order, formatting is lost, and this compact engine is less accurate than large cloud OCR services. It runs best in desktop Chrome or Edge.
FAQ
Is my image uploaded anywhere?
No. The image is processed locally by Tesseract.js in this tab. Only the engine and the English language data are downloaded, and they contain no information about you.
Why is the first run slower than later ones?
On the first visit the browser fetches roughly 10 MB of engine and language files. They are stored in the browser cache, so later runs start much faster unless you clear your cache.
How can I get better results?
Use a sharp, well-lit image with straight text filling most of the frame. Cropping unrelated areas and using a real screenshot both help.
More tools
2
Object Detector
Find common objects in a photo and draw labeled boxes around them, using a detection model that runs locally.
3
AI Image Describer
Get a short, plain-English sentence describing what appears in a photo, generated entirely on your own device.
4
Speech to Text
Transcribe English speech from an audio file or your microphone using a Whisper model that runs on your own device.
5
Text Summarizer
Condense a long English article or report into a few sentences, generated by a model that runs on your device.
6
Sentiment Analyzer
Paste English text and see whether each line reads as positive or negative, with a confidence score for each.