OCR
Extract text from images using optical character recognition.
About OCR
OCR extracts text from images using optical character recognition, turning a screenshot or photo of a document into editable, searchable text. Reach for it when you need to copy text out of a picture, a scanned page, or a screenshot you can't select. Recognition runs entirely in your browser, so the image you process never gets uploaded.
- Category
- export
- Input
- Accepts: image/jpeg, image/png, image/webp, image/tiff or image/bmp.
- Output
- Outputs: text/plain.
- Cost
- Free, runs in your browser
- Memory
- high
Common uses
- Pull the text out of a screenshot so you can paste it instead of retyping it
- Digitize a photographed receipt or printed page into plain text
- Recover wording from an image-only document that has no selectable text
- Grab a quote from a meme or infographic without manual transcription
- Extract a code snippet or error message captured in a screenshot
- Convert a scanned form (TIFF or BMP) into text you can edit and search
Frequently asked questions
Which image formats can it read?
JPEG, PNG, WebP, TIFF, and BMP. Clear, high-contrast images of printed text give the best results.
Does my image get uploaded?
No. The OCR runs in your browser, so screenshots, scans, and photos of sensitive documents stay on your device.
Will it read handwriting?
It's tuned for printed and typed text. For handwritten notes, Read Handwriting uses a vision model better suited to cursive and script.
Why isn't the text perfectly accurate?
OCR accuracy depends on resolution, contrast, fonts, and skew. Sharp, straight, well-lit images of standard fonts recognize far better than blurry or rotated ones.
What's the output format?
Plain text, which you can copy directly or chain into other text tools like word count, diff, or translate.
Keywords
- ocr
- text
- extract
- image
- recognition
- scan
- read