Hello,
I’m building a workflow to automate the digitization of handwritten invoices for my elderly relative, a trucker by trade. His clients are moving towards digital processes, and his handwritten invoices are becoming a barrier.
My goal is to:
- Monitor Google Drive: Detect new invoice PDF files. Done with a webhook and Drive Apps Script
- Check Processing Status: Verify if an invoice has already been processed . Done
- Send to Nanonets OCR: If unprocessed, send the PDF to Nanonets for optical character recognition.
- Retrieve Data: Get the extracted data from Nanonets.
- Populate Google Sheets: Fill a Google Sheet with the extracted data, effectively creating a digital version of the invoice.
I’ve successfully tested Nanonets and found its handwritten recognition to be excellent. However, I’m stuck at the integration stage, specifically:
- Google Drive Download to Nanonets HTTP Request: I can download the PDF from Google Drive, but I don’t know how to correctly configure the n8n HTTP Request node to send it to Nanonets.
- Nanonets File Upload: I’m struggling with the specific parameters and headers required to upload the PDF file to Nanonets via the HTTP Request node.
- Data Retrieval from Nanonets: I’m unsure how to parse and extract the OCR data returned by Nanonets.
Essentially, I need help bridging the gap between downloading the PDF from Google Drive and successfully utilizing the Nanonets API to extract and retrieve the invoice data.
I’m looking for assistance from individuals who have experience with:
- OCR workflows (specifically with handwritten documents).
- Nanonets API integration.
- n8n HTTP Request node configuration for file uploads.
- Google Drive and Google Sheets integrations.
While I’m currently using Nanonets due to its strong handwritten recognition, I’m open to exploring alternative OCR solutions if they offer better integration or ease of use.
Thank you for any guidance and support you can provide.