OCR-EDI

OCR for invoice & document EDI ingestion, AI & data integration applied to EDI.

Definition

EDI OCR extracts structured text from images or PDFs of invoices and commercial documents. Typical combinations: Tesseract (open-source), Google Document AI, AWS Textract, Azure Form Recognizer. Often paired with a downstream LLM to transform raw text into EDI JSON (Invoice, ASN, Receipt). Error rate < 1% on standard printed invoices.

Origin

OCR discipline since the 1950s, transformed by Convolutional Neural Networks (2012+), industrialised by Google Document AI (2020) and Azure Form Recognizer (2019).

Example in context

OCR pipeline: PDF invoice → Tesseract / Document AI → raw JSON → LLM normalisation → signed UBL Invoice.

Related terms

• Document AI — standardised GCP service.