Skip to content

Usage in n8n

Once you have purchased PlemeoOCR and obtained your API key, you can integrate OCR functionality into your n8n workflows.

Add the Node

Steps to Add PlemeoOCR Node

  1. Open your n8n editor
  2. Add a new node to your workflow
  3. In the search bar, type "plemeo OCR" and select the node

Add OCR NodeAdding the PlemeoOCR node in n8n

Node Location

The PlemeoOCR node will appear in your n8n node library under:

  • Custom nodes (if installed as a community node)
  • API/HTTP nodes section
  • Search results when typing "plemeo" or "OCR"

Configure the Node

Preset Configuration

The following parameters are preset and do not need to be changed:

ParameterValueDescription
API Hostnameai-test.plemeo.deServer address for OCR processing
API Port8000Communication port
API Path/process_ocrEndpoint for OCR requests
Use HTTPDeactivatedUses HTTPS by default for security

Node ConfigurationPlemeoOCR node configuration panel

Required Configuration

You must configure the following:

API Key

  • Field: API Key
  • Value: Your personal API Key from the app description
  • Location: Found in plemeo.ai → My Apps → PlemeoOCR

Configuration Steps

  1. Select the PlemeoOCR node in your workflow
  2. Open the node configuration panel
  3. Verify preset values (hostname, port, path)
  4. Enter your API Key in the designated field
  5. Save the configuration

Integration with Other Nodes

Common Workflow Patterns

1. File Input → OCR → Text Processing

[File Upload] → [PlemeoOCR] → [Text Analysis]

2. Cloud Storage → OCR → Knowledge Base

[Dropbox/Google Drive] → [PlemeoOCR] → [Database Storage]

3. Email Attachments → OCR → Automation

[Email Trigger] → [Extract Attachments] → [PlemeoOCR] → [Process Data]

Compatible Nodes

PlemeoOCR works well with:

  • File Input Nodes: Dropbox, Google Drive, FTP, HTTP Request
  • Processing Nodes: Text manipulation, data transformation
  • Output Nodes: Database storage, email notifications, webhooks
  • Conditional Nodes: Switch, IF statements based on OCR results

Automatic Knowledge Base Storage

Background Processing

When PlemeoOCR processes a document, it automatically:

  1. Extracts text from the input document
  2. Stores the text in your plemeo.ai Knowledge Base
  3. Makes it searchable through the AI chatbot
  4. Returns the text to your n8n workflow for further processing

Best Practices

File Optimization

  • Image Quality: Use high-resolution images for better OCR accuracy
  • File Size: Optimize file sizes for faster processing
  • Format Selection: PDF and PNG typically provide best results

Error Handling

  • Add error handling nodes after PlemeoOCR
  • Check confidence scores before using extracted text
  • Implement retry logic for failed processing

Performance

  • Batch Processing: Process multiple files efficiently
  • Async Operations: Use n8n's async capabilities for large files
  • Monitoring: Track processing times and success rates

What's Next?