OCR: recognise text
Turn a scan or a photo of a document into text you can copy, edit or search, without retyping it.
- No file uploads
- Free, no sign-up
- Works in your browser
How to recognise text with OCR in 3 steps
- Choose a scanned PDF or an image (JPG, PNG or WebP).
- Pick the document's language: Spanish, English or both.
- Press "Recognise text": copy the result or download it as .txt or as a searchable PDF.
Your files never leave your device
Unlike other online tools, DocuSmart doesn't upload your documents to any server: everything is processed inside your browser. You can use it with contracts, invoices or ID documents with complete peace of mind. Read our privacy policy (in Spanish).
Frequently asked questions
What is a searchable PDF?
It's the same scanned document with an invisible layer of text on top: it looks the same, but you can select, copy and search for words.
Are my documents uploaded to the internet?
No. Recognition is done with Tesseract inside your browser. The first time, the language model is downloaded from docusmart.online and stays saved in your browser for next time.
How accurate is it?
With sharp, straight scans, very accurate. Blurry, crooked or poorly lit photos lower the quality: DocuSmart warns you if the result is doubtful.
Does it recognise handwriting?
It works with printed text. Handwriting is usually recognised poorly.
More free tools
Take DocuSmart on your phone
The Android app adds a scanner with edge detection, a PIN-protected secure folder, Word, Excel and PowerPoint, and OCR. It works offline.