International

OCR + extraction

Read the page before you analyse it

Scanned PDFs and images need OCR. The app tells you when a file looks like a scan instead of pretending analysis succeeded.

Demonstration

Scan a page

  1. 1. Capture or import a page
  2. 2. Review and rotate if needed
  3. 3. Extract text in the installed app
  4. 4. Ask Anything about the result

Camera permission is requested only when you scan. This website does not OCR your photos.

The problem

A scan without text cannot be searched or cited honestly.

The solution

OCR produces searchable text. Analysis, terms and Q&A then run on that text. Limitations are shown.

How it works

1. Open the scan

PNG, JPG, TIFF or a scanned PDF.

2. Run OCR

Bundled Tesseract is used on this computer.

3. Continue

Ask questions or extract terms from the recognised text.

Capabilities

Evidence

Substantive findings should show a source location and an excerpt from the imported file. If the file does not contain the answer, the software says it was not identified. It does not invent clauses, parties, dates or amounts.

Privacy

Local AI keeps document work on this device when a tested local model is active. Optional cloud AI is off unless you approve sending excerpts to a provider you configure. See local AI setup and privacy.

Related

FAQ

Can it process scanned documents?

Yes, for supported image and scan formats. OCR quality depends on the page. Warnings are shown when extraction is limited.

ContractAnalysis.ai is document-processing and analysis software. It does not provide legal advice and does not replace a qualified lawyer, solicitor, attorney or other professional adviser. AI-generated results may contain errors and should be independently reviewed before being relied upon.