Skip to main content

Documentation Index

Fetch the complete documentation index at: https://hedra.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

n8n is a workflow automation tool that lets you connect APIs visually. This guide shows how to use n8n’s HTTP Request node to automate content generation with the Hedra API.

​
Authentication setup

Before building workflows, create a reusable credential for the Hedra API:
  1. Go to Credentials in n8n
  2. Click Add Credential → Header Auth
  3. Set Name to X-API-Key and Value to your Hedra API key
  4. Save as “Hedra API Key”
Use this credential in every HTTP Request node by selecting Authentication → Generic Credential Type → Header Auth → Hedra API Key.

​
Example: Generate an image on a schedule

This workflow generates a daily image from a prompt.

​
Node 1: Schedule Trigger

  • Type: Schedule Trigger
  • Interval: Every day at 9:00 AM

​
Node 2: Generate Image

  • Type: HTTP Request
  • Method: POST
  • URL: https://api.hedra.com/web-app/public/generations
  • Authentication: Header Auth → Hedra API Key
  • Body (JSON):
{
  "type": "image",
  "text_prompt": "A vibrant abstract background for social media",
  "ai_model_id": "a66300b4-f76e-4c4a-ac41-b31694ff585e",
  "aspect_ratio": "1:1",
  "resolution": "1080p"
}

​
Node 3: Wait

  • Type: Wait
  • Duration: 30 seconds

​
Node 4: Check Status

  • Type: HTTP Request
  • Method: GET
  • URL: https://api.hedra.com/web-app/public/generations/{{ $('Generate Image').item.json.id }}/status
  • Authentication: Header Auth → Hedra API Key

​
Node 5: Check if Complete

  • Type: If
  • Condition: {{ $json.status }} equals complete
  • True: Continue to next node
  • False: Loop back to Wait node

​
Example: Avatar video from text input

This workflow takes a text input, generates speech, and creates an avatar video.

​
Node 1: Manual Trigger / Webhook

Start the workflow manually or via a webhook with a text field in the body.

​
Node 2: Upload Portrait Image

If you have a fixed portrait, upload it once and reuse the asset ID. Otherwise, create and upload:
  • Type: HTTP Request
  • Method: POST
  • URL: https://api.hedra.com/web-app/public/assets
  • Body (JSON):
{
  "name": "portrait.png",
  "type": "image"
}

​
Node 3: Upload File

  • Type: HTTP Request
  • Method: POST
  • URL: https://api.hedra.com/web-app/public/assets/{{ $json.id }}/upload
  • Authentication: Header Auth → Hedra API Key
  • Body: Form-Data, attach your portrait image file

​
Node 4: Generate Avatar Video with Inline TTS

Use audio_generation to skip a separate TTS step:
  • Type: HTTP Request
  • Method: POST
  • URL: https://api.hedra.com/web-app/public/generations
  • Body (JSON):
{
  "type": "video",
  "ai_model_id": "26f0fc66-152b-40ab-abed-76c43df99bc8",
  "start_keyframe_id": "{{ $('Upload Portrait Image').item.json.id }}",
  "audio_generation": {
    "type": "text_to_speech",
    "voice_id": "f412c62f-e94f-41c0-bfc6-97f63289941c",
    "text": "{{ $('Manual Trigger').item.json.text }}"
  },
  "generated_video_inputs": {
    "text_prompt": "A person speaking to the camera",
    "aspect_ratio": "9:16",
    "resolution": "540p"
  }
}

​
Node 5: Poll Until Complete

Use a Wait + HTTP Request + If loop (same pattern as the image example above) to poll GET /generations/{id}/status until status is "complete".

​
Polling pattern

Most Hedra generations are asynchronous. Use this reusable polling pattern in your n8n workflows: 
[Generate] → [Wait 15-30s] → [Check Status] → [If complete?]
                  ↑                                  │
                  └──────── No ──────────────────────┘
                                                     │
                                              Yes → [Next step]
For short generations (images, TTS), a 15-second wait is usually enough. For long-form avatar videos, use 60-second intervals and check the progress field (0-1) in the status response.

​
Tips

  • Reuse asset IDs. If you use the same portrait or audio across runs, upload once and store the asset_id for reuse.
  • Use n8n variables. Store your Hedra model IDs and voice IDs as n8n workflow variables for easy updates.
  • Error handling. Add an If node to check for status equal to "error" and branch to a notification or retry path.