Novobi OCR - Azure Document Intelligence
by Novobi https://www.novobi.com , Inc https://www.novobi.com$ 9999.99
In-App Purchases| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
•
Invoicing (account)
• Discuss (mail) |
| Community Apps Dependencies | Show |
| Lines of code | 3326 |
| Technical Name |
nvb_ocr_azure |
| License | LGPL-3 |
| Website | https://www.novobi.com |
| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
•
Invoicing (account)
• Discuss (mail) |
| Community Apps Dependencies | Show |
| Lines of code | 3326 |
| Technical Name |
nvb_ocr_azure |
| License | LGPL-3 |
| Website | https://www.novobi.com |
Novobi OCR – Azure Document Intelligence
Microsoft Azure AI-powered document extraction for Odoo
Enterprise-Grade OCR, Powered by Azure
This module connects the Novobi OCR framework to Microsoft Azure Document Intelligence (formerly Form Recognizer), Microsoft's enterprise AI service for extracting structured data from invoices, receipts, and custom document types — with industry-leading accuracy.
Azure Integration Features
Prebuilt Invoice Model
Uses Azure's prebuilt-invoice model out of the box.
No training required — works on vendor invoices from day one.
Custom Model Support
Configure a custom Azure model ID per provider for specialized document layouts or industry-specific invoice formats.
Async Polling
Submits documents asynchronously and polls the Azure operation URL until results are ready — up to 120 seconds, with retries.
Extracted Fields
Azure's prebuilt-invoice model extracts all the fields you need
Vendor
- Vendor name
- VAT / tax ID
- Vendor address
Invoice Header
- Invoice number
- Invoice date & due date
- Purchase order number
- Currency code
- Payment terms reference
Line Items
- Description & product code
- Quantity & unit price
- Line subtotal
- Tax amount & tax rate %
- Totals: subtotal, tax, total
Intelligent Data Parsing
-
Multi-format date parsing
Handles ISO 8601, US (MM/DD/YYYY), EU (DD/MM/YYYY), and written month name formats — automatically normalized to YYYY-MM-DD. -
Multi-locale amount parsing
Parses US format (1,234.56), EU format (1.234,56), and plain numbers. Also handles currency symbols and whitespace. -
Tax rate extraction
Reads"18 %",0.18, or18from Azure and normalizes all to a percentage for Odoo'saccount.taxmatching.
-
Smart price resolution
When unit price is missing, computes it from line total / quantity. Supports negative amounts for credits and discounts. -
Bracket SKU parsing
Treats[SKU123] Product Nameformat to split vendor code from display name for more accurate product matching. -
Vendor matching hierarchy
Matches vendors by VAT → IBAN + name → fuzzy name (SequenceMatcher). VAT normalization handles country-code prefixes and punctuation.
Configuration
- Go to Settings → General Settings → Document Intelligence (OCR).
- Set Vendor Bills OCR Provider to Azure Document Intelligence.
- Open the provider configuration and enter your Azure Endpoint URL and API Key.
- Optionally, enter a custom Model ID (leave blank to use
prebuilt-invoice). - Set the state to Enabled (or Test for staging).
- Set your desired Confidence Threshold (default: 80%).
Supported File Types
Requirements
- Requires Novobi OCR Base
- Requires Novobi OCR – Vendor Bill Mapping
- An active Microsoft Azure subscription with Document Intelligence enabled
- Python package:
requests
Please log in to comment on this module