Free Online OCR — PDF & Image to Word
Turn scanned pages into editable text. Review the results, then download a Microsoft Word document.
01 Add your documents
Up to 20 pages per batchScans, photographs, or PDFs with selectable text
PDF, JPG, PNG, WebP, BMP · 20 MB per file10 files / 50 MB total · clear printed text works best
02 Extraction settings
For a scan with an incomplete or incorrect text layer, choose “OCR every selected page”. Images always use OCR.
Choose a file to get started.
Your files and extracted text stay in this browser tab. This tool does not upload them to a server. The first OCR run loads the recognition engine and language files.
03 Review & download
0 pagesChoose a document and select Extract text.
Check names, numbers, and punctuation before use.
Compare with source page
Word downloads join wrapped scan lines into paragraphs and can retain simple detected columns. Alignment is adjustable. Tables, page borders, images, handwriting, and complex layouts may still need manual correction.
Free online OCR for PDFs and images
Optical character recognition (OCR) converts text in a scanned document or photograph into text you can edit, copy, and reuse. Use this PDF and image to Word tool for reading materials, printed notes, library documents, and other clear printed pages.
How to convert a PDF or image to Word
- Choose your files. Add a PDF, JPG, PNG, WebP, or BMP image.
- Select the text language. Choose English, Bengali, or both. For a large PDF, enter the pages you need.
- Extract and review. Select Extract text, choose each result page, and correct any recognition errors.
- Download your document. Select Download Word for an editable .docx file, or download plain text.
Get clearer OCR results
- Use a sharp, well-lit image with the page lying flat.
- Rotate sideways pages before recognition using the rotation option.
- Try 300 dpi for small printed text in a PDF.
- Select the language that matches the document.
- Proofread the result, especially names, dates, and numbers.
Frequently asked questions
Can I convert a scanned PDF to editable Word?
Yes. The tool renders selected PDF pages as images and recognizes their printed text. Wrapped lines are joined into paragraphs; simple detected columns can be retained in Word. You can edit the text and choose alignment before downloading. Images and exact page decoration are not reproduced.
Which image formats are supported?
You can use JPG, JPEG, PNG, WebP, and BMP. Convert TIFF, HEIC, or other formats to JPG or PNG before adding them. Each file can be up to 20 MB, with a maximum of 10 files, 50 MB, and 20 pages per batch.
Does this OCR tool support Bengali?
Yes. Choose Bengali / বাংলা for Bengali printed text, or English + Bengali for a document containing both. Accuracy depends on the typeface, image quality, and page layout.
Are my documents uploaded or stored?
This tool processes documents in your browser and does not send the files or extracted text to a server. Recognition engines and language files are loaded from this website; language data may be cached by your browser. Closing or refreshing the tab discards your work. Other website plugins, analytics, and browser extensions have their own behavior.
What is the difference between Auto and OCR every page?
Auto uses a PDF’s existing text layer when it contains enough usable text; otherwise it recognizes the rendered page. If a PDF has only a partial text layer or its text is incorrect, choose OCR every selected page. Images always use OCR.
How can I improve the Word layout?
Use Detected paragraphs & columns for OCR results, or choose Flowing paragraphs for a single-column document. Text corrections keep their column placement when the detected paragraph count is unchanged. Adding or removing paragraphs switches that page to single-column output to include all edits safely. Select Justified, Left, Centered, or Right alignment if needed. PDF text-layer extraction uses paragraph formatting; choose OCR every selected page to detect its columns.
Will the Word download preserve tables and handwriting?
The download contains editable paragraphs and can retain simple OCR-detected columns. It does not reconstruct table cells, images, page borders, or exact page design. This tool is intended for printed text; handwriting and complex formulas may not be recognized reliably.
Why does the first extraction take longer?
Your browser first loads the OCR engine and selected language data. Large scans and slower devices also take longer. Keep this tab open while processing, and use smaller batches on a phone.
Can I use password-protected PDFs?
Password-protected PDFs are not supported by this tool. Create an unlocked copy that you are authorized to use, then select the required pages.