On this page
Azure Document Intelligence: document processing features and setup
A Microsoft service for extracting text, tables, and fields from documents. The website presents it as part of the Content Understanding and Foundry environment, so it is important to choose the interface and model that suit the application.
Tool by the Otomatzia editorial team · source content checked on
Overview
Choosing between a prebuilt model and custom extraction in Document Intelligence starts with the structure of the documents the business receives. An invoice model is not necessarily suitable for an internal form, and general text extraction does not always provide the meaning of each field. Because product names and their placement in Microsoft's environment can change, document the endpoint, API version, and model used to test the workflow.
Features
- Extracts structure and fields through a REST API
- Prebuilt models for document types
- Custom extraction for unique forms
Business uses
- Transfer form data to a business system after validation
- Extract an invoice table for review
Getting started
- Choose a model and an Azure environment, then test a familiar document before connecting a permanent workflow.
- Run a typical document and one that deviates from the template through the same model. Compare required fields, table structure, and missing values before deciding whether customization is needed.
- Test a REST request from the intended application environment, and record the region, permissions, and API version alongside the expected output so you can detect behavior changes on the next connection.
What to check
- Check service names, API versions, and regional availability during setup.
- Check Hebrew support and table structure in the specific model, including missing fields.
Sources to check next
Vendor features and prices change. Confirm details on the linked official pages before you rely on them.