Azure Document Intelligence

Microsoft's document OCR — steady layout and tables, 500 free pages a month.

Access Azure Portal

1. 01

   Search & Open Azure Portal

   Search Microsoft Azure Portal in your browser, or visit:
   https://azure.microsoft.com/en-us/get-started/azure-portal
   
   ⚠ Users in mainland China should use a VPN — the domestic CN Azure has different services.

   
2. 02

   Azure Homepage

   On the Azure homepage, click Sign In at the top-right to log in, or click to register a free account.

   

Sign In or Register

1. 03

   Sign In With Microsoft Account

   Sign in with your existing Microsoft or GitHub account, or register a new one.
   New Azure Free Account users typically receive USD 200 credit, valid for 30 days.

   
2. 03A

   Register Free Account (Optional)

   To register a free account, visit:
   https://azure.microsoft.com/en-us/pricing/purchase-options/azure-account
   You'll need: Microsoft/GitHub account + phone number + credit or debit card (verification only, not charged during free period).

   
3. 03B

   Registration Complete

   After registration you'll be redirected to the Azure Portal home. Existing accounts land on the same page after signing in.

   

Create Document Intelligence Service

1. 04

   Search "Document Intelligence"

   In the Portal search bar, type:
   • English: Document Intelligence
   • Chinese UI: 文档智能
   
   ⚠ Current OCR is under Document Intelligence — NOT the old Computer Vision service.

   
2. 04A

   Document Intelligence Service Page

   After searching, you'll land on the Document Intelligence page. Click + Create to start creating a new service resource.

   
3. 04B

   Fill in the Creation Form

   Fill in these fields (keep others as default):
   • Subscription: Select your subscription
   • Resource Group: Create new or select existing
   • Region: Choose the region closest to you
   • Pricing Tier:
   — F0 (Free): 500 pages/month limit, ideal for personal evaluation, no charge
   — S0 (Standard): No page limit, pay-per-use, for regular production use
   Recommended: start with F0 to try it out, upgrade to S0 when needed

   
4. 04C

   Review & Deploy

   Confirm your Pricing Tier selection (F0 free / S0 standard), then click Review + Create → Create. Wait 1–3 minutes for deployment.

   

Path A · Online Test (Optional)

1. 05A

   Enter the Online Test UI

   After deployment, click Go to resource to open the service detail page.
   Then click "Try in Document Intelligence Studio" to open the online test environment.

   
2. 05B

   Redirect to Studio

   Inside Document Intelligence Studio, choose the OCR-related module on the left panel.

   
3. 05C

   Select OCR / Read Function

   From the function list, select Read (general OCR) for recognizing text in images or PDFs.

   
4. 05D

   Select Your Service Configuration

   In the dropdown menus, select the subscription and resource you just created to link the test with your service.

   
5. 05E

   Upload File & Test

   Drag or upload a local image or PDF in the top-left area, then click Run analysis to start recognition.

   
6. 05F

   View Recognition Results

   Recognition results appear on the right. View specific fields or export results in JSON format.

   

Path B · Get API Key & Endpoint (Required)

1. 06

   Copy API Key & Endpoint

   Back on the service detail page, click Keys and Endpoint in the left menu.
   Copy Key 1 (or Key 2) and the Endpoint URL. Save securely — never share them.
   Both are required to connect Lectio to your Azure service.

   

Configure Lectio App

1. 07A

   Open Engine Management

   Open the Lectio app, go to Settings → Engine Management to see the list of OCR engines.

   
2. 07B

   Select Microsoft Azure

   Find Microsoft Azure in the engine list and tap Add / Configure to open the setup screen.

   
3. 07C

   Enter Credentials & Save

   Enter the API Key and Endpoint from Step 06 into the respective fields.
   Tap Test to verify connection, then tap Save to complete setup.