Image to Text (OCR)
Read text out of a photo, screenshot or scan.
Runs entirely in this tab. Nothing is uploaded.
Drop an image here, or click to choose
Photos of documents, screenshots and scans
The recognition model (a few MB) downloads once from a CDN and is then cached. Your image is never part of that request — it stays in this tab.
Frequently asked questions
Is my image uploaded?+
No. The recognition model is downloaded to your browser and runs here; the image is never part of any request. This is the one tool on the site that fetches something (the model, a few MB, cached after the first run).
How do I get better results?+
Use the sharpest, most straight-on image you have, with good contrast and the text upright. Photos taken at an angle or in low light lose accuracy quickly.
Which languages work?+
English, French, German, Spanish, Italian, Portuguese and Dutch are selectable. Each downloads its own model the first time you use it.
Tools that pair with Image to Text (OCR)
- PDF to TextPull the text layer out of a PDF and copy or download it.
- Bank Statement to ExcelPull transactions into a spreadsheet, and check every row adds up.
- PDF to WordRebuild a PDF as an editable Word document, not floating text boxes.
- Word to PDFConvert a .docx to PDF through your browser's own print engine.
- Markdown ↔ HTMLConvert Markdown to HTML and back, live as you type.
- Document ConverterConvert text documents between TXT, HTML and Markdown.