How to make a scanned PDF searchable
- Add a scanned PDF, or a photo of a page as JPG or PNG.
- Choose what you want back: a searchable PDF that looks identical but can be searched and copied from, or plain text.
- Pick a detail level. 200 dpi suits ordinary office scans; choose 300 for small print.
- Press Read the text. The first run downloads the recognition engine, about 6 MB, which your browser then keeps.
About this tool
OCR — optical character recognition — finds the words in a picture of a page and turns them into real text. This free OCR tool runs the recognition on your own device, which is unusual, because recognition is expensive and almost every free service therefore does it on their servers.
That difference matters more here than anywhere else on this site. Think about what people actually scan: passports and visas, payslips, tax returns, medical letters, signed contracts, bank statements for a mortgage application. Uploading those to a free website to have the text pulled out means handing over exactly the documents you would least like copied. Tesseract, the open-source recognition engine, compiles to WebAssembly, so the whole process happens in the tab instead. The honest cost is the download: the engine and the English language data come to roughly 6 MB the first time, after which your browser caches them.
A searchable PDF is usually what you want. It keeps the scan looking exactly as it did and adds an invisible text layer positioned over the image, so the document can be searched, copied from and indexed while still looking like the original. Recognition quality depends heavily on the scan: sharp, straight, well-lit pages at 300 dpi read almost perfectly, while a dim phone photo at an angle will not. This engine reads the Latin alphabet in English; other alphabets need their own language data, which is not bundled here.
Frequently asked questions
- Is my scan uploaded for recognition?
- No, and that is the reason this tool exists. Recognition runs on your device using WebAssembly. The counter at the top of the page stays at zero bytes.
- Why is there a 6 MB download the first time?
- That is the recognition engine and the English language data. Services that avoid it do so by sending your document to their servers instead. Your browser caches it, so it happens once.
- What is a searchable PDF?
- The original scan with an invisible layer of recognised text positioned exactly over the words in the image. It looks unchanged but can be searched and copied from.
- Which languages are supported?
- English, using the Latin alphabet. Other languages need their own trained data, which is not included in this build.
- How do I get the best results?
- Scan straight, in good light, at 300 dpi, in black and white or greyscale. Skew, shadows and low resolution cost far more accuracy than anything else.