Extract Invoice Data with OCR, Gemini AI & Airtable

This n8n workflow revolutionizes invoice processing by automating data extraction from various file formats (PDF, JPG, PNG). It triggers automatically when files are dropped into a monitored folder, uses Tesseract OCR via community nodes to detect and extract raw text, and leverages Google Gemini AI to intelligently parse and structure critical fields like invoice number, date (standardized to YYYY-MM-DD), subtotal, tax, total, currency, vendor name, email, phone, and ship-to address.

The extracted data is seamlessly sent to Airtable for organized storage and easy querying, eliminating manual entry errors and saving hours of tedious work. Optional Telegram integration notifies users instantly upon successful processing, ensuring real-time updates. Ideal for freelancers, small businesses, accountants, and finance teams handling high volumes of invoices.

Designed for self-hosted n8n with community nodes (n8n-nodes-tesseractjs, n8n-nodes-pdf-page-extract), it requires Docker setup, folder mounting, and credential configuration for Gemini API, Airtable, and Telegram. This lightweight solution scales effortlessly, reduces operational costs, and integrates AI-powered accuracy into everyday workflows.