How to extract text from scanned PDFs and images using OCR.
The Extract Text (OCR) tool uses Optical Character Recognition to extract text from scanned PDFs and image-based documents.
## When to Use OCR
Use OCR when:
- You have a scanned PDF (photos of paper)
- The PDF doesn't allow text selection or copying
- You need to extract data from an image
Regular digital PDFs (created from Word, InDesign, etc.) don't need OCR — use a regular PDF viewer.
## Supported Input
- PDF files (scanned)
- PNG, JPG, TIFF images containing text
## Output Options
- **Plain Text** — Raw extracted text, download as .txt
- **Searchable PDF** — Original PDF with an invisible text layer added
- **Word Document** — Extracted text formatted in .docx
## Language Support
OCR supports 100+ languages. Select your document language before processing for best accuracy.
## Accuracy Tips
- Higher resolution scans produce better results (300+ DPI)
- Black and white documents are more accurate than color
- Clean, printed text is more accurate than handwriting
- The tool cannot handle handwriting (use AI Describe Image for that)
## Privacy
OCR uses WebAssembly (Tesseract.js) and runs entirely in your browser. Your documents are never uploaded.
