Invoice Currency Formatting Tool
Format an amount the way its own country writes it, and spell it out in words.
Read scanned invoices into structured JSON, with a confidence figure on every field.
Drop scans, photos or PDFs here — several at once
Invoice OCR to JSON reads scanned invoices into structured JSON with real types and ISO dates, and reports for every field whether it was read from a label, inferred or missing — so doubtful invoices can be routed to a person.
If you are automating anything, JSON is the shape you want an invoice in: nested, typed, and ready to post to an API. The extraction services that produce it charge per page and want the document on their servers.
This produces the same structure in your browser. Drop scans or PDFs and each comes back as a JSON object: the invoice header, both parties, the tax breakdown, the totals and an array of line items, with numbers as numbers and dates as ISO 8601 strings.
Crucially, it also reports how it knows. Each field says whether it was read from a label, inferred, or missing, and OCR pages carry their confidence — so your own code can route the doubtful ones to a human instead of trusting everything equally. A JSON Schema is available to validate against.
Numbers as numbers, dates as ISO 8601.
Labelled, inferred or missing, per field.
A folder of scans as one JSON array.
Validate the output before you trust it.
Unlike every extraction API.
Scans, photos or PDFs.
One object per invoice, or an array for a batch.
Every field says how it was found.
Pretty-printed or minified.
For OCR it is the engine’s own certainty about the characters, 0 to 100. For fields it is whether the value came from a label, from its position, or from a guess — which is the more useful signal.
Yes. The JSON Invoice Generator publishes a JSON Schema for this shape, and the Invoice JSON Validator checks a file against it.
No — it runs in the page. For an API you would run something server-side; this is for the cases where the document must not leave the machine.
Numbers, with the currency as a separate ISO code, so no parsing is needed on the other side.
What people who use the Invoice OCR to JSON think of it. Reviews come from visitors and are posted as written.
No ratings yet
Yours would be the first.
0 reviews
Newest first
No reviews yet
Be the first to rate the Invoice OCR to JSON. It takes half a minute and no account.
Write the first reviewFormat an amount the way its own country writes it, and spell it out in words.
Check a billing address is complete and correctly formed for its country, before it is invoiced.
Mark an invoice as past due so it cannot be mistaken for the original.
Split a PDF of many invoices into one file each — including where each invoice starts.
Download a real .docx invoice you can edit in Word, Docs or LibreOffice.
Reconcile your ledger against a supplier statement or a bank export.
Parse an invoice into structured fields and see exactly how each one was found.
Check any tax number on an invoice — VAT, GSTIN, ABN, EIN, PAN or IBAN — against its own rule.