Turn a photo, screenshot or scanned page into editable text. Useful for site drawings, material quotations and invoices that only exist as images or scans. It all runs in your browser.
Recognition happens on your own device, so your file is not sent to BuildCostMate. Language data is downloaded once from a public CDN the first time you use a language.
Which files work best?
Sharp, well-lit, straight-on images at 200â300 DPI give the most accurate text. Blurry or skewed photos reduce accuracy.
Can it read handwriting?
Not reliably. It is built for printed or typed text.
My PDF already has selectable text. Do I need OCR?
No. Use PDF to Word or PDF to Excel for text-based PDFs; they are faster and more accurate.
Is there a page limit?
Scanned PDFs are read up to 30 pages per run to keep your browser responsive.
Choose a photo, screenshot or scanned PDF and a language. The tool finds the ruling lines of any table, reads each cell on its own, reads the rest of the page, and then rebuilds font sizes, bold, colours and positions. You can copy the text, download a text file, or download a Word or Excel file that keeps the layout. Everything runs in your browser, so your document is not uploaded.
More tips are in our guide How to convert a scanned PDF or photo to Word with OCR.
Does OCR work on handwriting?
Not reliably. It is designed for printed or typed text.
Are my files uploaded?
No. Conversion runs in your browser.
Which languages are supported?
English, Urdu, Hindi, Arabic, French, German and Spanish. Language data downloads the first time you use a language.