The Document Recognition workflow action in Yeeflow is a powerful tool that leverages Microsoft Azure AI Document Intelligence to extract data from documents, such as PDF files and images, and return the recognized data for further processing. This action can either use Yeeflow’s built-in Azure AI Document Intelligence service or connect to your own Azure AI service if configured.
When to Use the Document Recognition Action
The Document Recognition action is ideal for automating data extraction tasks for various business needs, including:
Invoice recognition
Automate the extraction of key data from invoices, including vendor details, invoice numbers, dates, line items, taxes, and totals. This streamlines accounts payable workflows, reduces errors, and accelerates payment processes.
Business card scanning
Quickly extract and organize contact details from business cards, such as names, job titles, companies, and contact information. This simplifies CRM data entry, boosts networking efficiency, and ensures accurate contact records.
Processing scanned documents
Digitize and extract data from receipts, contracts, and letters to make them searchable and actionable. This reduces manual review, accelerates document handling, and ensures accurate recordkeeping.
Extracting data from forms and agreements
Extract information from application forms, surveys, and agreements, including names, dates, and terms. This speeds up form processing, ensures accurate data capture, and supports compliance and operations workflows.
By using this action, you can streamline workflows, reduce manual data entry, and enhance process efficiency.
Configuring the Document Recognition Action in Workflow Designer
Follow these steps to configure the Document Recognition action:
1. Action Name: Enter a name for the action to identify its purpose in the workflow.
2. Description: Add a brief description to explain the action’s functionality or purpose.
3. Select Input File: Choose an attachment-type variable from your workflow to serve as the input file. Alternatively, use the Expression Editor to select a dynamic value as the input file.
4. Select Provider: Choose the service provider for the document recognition:
Built-in Service: Leverages Yeeflow’s built-in Azure AI Document Intelligence service.
Azure AI Document Intelligence: Uses your own Azure AI service configured in your Azure account.
5. Select Connection (Only for Azure AI Document Intelligence): This dropdown appears when you select “Azure AI Document Intelligence” as the provider. Choose a pre-configured connection from the list.
Learn more about connections in the Introduction to the Connections Function in Yeeflow Admin Center.
6. Processing Models: Select a processing model from the dropdown list. The available models depend on the selected Azure AI service.
Learn more about Azure AI Document Intelligence processing models by visiting Azure AI Document Intelligence Model Overview.
7. Page Range: Define the page range for data recognition:
All pages: Recognizes data from the entire document.
Range: Specify a range of pages for recognition.
8. Save Output To: Choose a variable to store the output content from the document recognition action. The return value will be in JSON format, containing a full result from Azure AI Document Intelligence.
9. Set Values to Variables: To extract and save specific fields from the returned JSON to individual variables:
Click the Select Fields button under “Set values to variables.”
A pop-up window will display the fields recognized by the selected processing model:
Field Name: The first column lists all recognized fields based on the processing model.
Field Type: The second column displays the type of each field (e.g., string, date, number).
Associated Variable: In the third column, choose a variable in your workflow to save the value of each field.
Use the checkboxes to select the fields you wish to map to variables.
With the Document Recognition action, you can automate data extraction tasks and seamlessly integrate them into your workflows. Whether you’re processing invoices, forms, or other documents, this action enhances efficiency and reduces manual effort. For more information on creating connections for Azure AI Document Intelligence, refer to the How to Add an Azure AI Document Intelligence Connection guide.