Auto-Generate Blog & AI Image from YouTube Videos with Dumpling AI & GPT-4o

📋 High-Level Summary

This workflow automates the entire process of converting a newly uploaded video (e.g., a YouTube video) stored in a Google Drive folder into a fully written blog post and a matching AI-generated image. It leverages Dumpling AI for video transcription and image generation, and OpenAI's GPT-4o for content creation. The final blog post and image are saved to an Airtable base, making it ideal for content creators who want to repurpose video content into blog articles with custom visuals without manual effort.

The workflow is triggered the moment a new video file appears in a specific Google Drive folder. It downloads the video, transcribes it, generates a blog post and an image prompt using AI, creates the image, and stores everything in Airtable. This saves hours of manual transcription, writing, and image design work.

🔧 Node-by-Node Breakdown

• Watch Folder for New YouTube Videos — Google Drive Trigger (OAuth2). This node monitors a specific Google Drive folder for newly created files. It polls every minute and triggers when a new file appears. You must configure it to watch a specific folder by selecting the folder from your Google Drive. Auth method: OAuth2 connection to Google Drive.
• Download Video File — Google Drive (OAuth2). Downloads the file that was just detected by the trigger. It uses the file ID from the previous node's output ($json.id). Operation set to "download". Auth: OAuth2 (same Google Drive credential).
• Convert Downloaded Video to Base64 — Extract From File (No auth). Converts the binary video file into a base64-encoded string so it can be sent to an external API. The operation is "binaryToProperty" which extracts the binary data into a property for the next node.
• Transcribe Video with Dumpling AI (Full Transcript) — HTTP Request (Header Auth). Sends a POST request to Dumpling AI's extract-video endpoint with the base64 video and a detailed transcription prompt. The prompt instructs the AI to produce a complete, accurate transcript including all spoken words. Auth method: Header Auth — you need to provide an API key as a header (typically Authorization: Bearer YOUR_KEY).
• Generate Blog & Image Prompt using GPT-4o — OpenAI (API Key). Uses GPT-4o to generate a blog post and an image prompt from the transcript received from Dumpling AI. The system message contains detailed instructions on how to write the blog post (conversational tone, structure, length) and how to create a descriptive image prompt suitable for FLUX.1-pro. The output is set to JSON format. Auth: API Key from your OpenAI account.
• Generate Visual from Blog Prompt with Dumpling AI — HTTP Request (Header Auth). Sends a POST request to Dumpling AI's generate-ai-image endpoint with the image prompt extracted from GPT-4o's output. Parameters include model FLUX.1-pro, image dimensions 1024x576, and aspect ratio 16:9. Auth: Header Auth (same Dumpling AI API key).
• Save Blog Post to Airtable — Airtable (API Key). Creates a new record in an Airtable table with the blog post content. The column mapping maps the GPT-4o generated blog content to the "Content" field in Airtable. Auth: API Key from your Airtable account.
• Upload Blog post Image to Airtable — HTTP Request (Header Auth). Patches the Airtable record (using its ID) to attach the generated image URL (which comes from the image generation node). Note: The URL used in the JSON body currently uses the image prompt text rather than the actual generated image URL — this may require adjustment to use the correct output from the Dumpling AI image generation node. Auth: Header Auth (Airtable API key).

🚀 Setup Instructions

To use this workflow, you need accounts and API keys for the following services:

1. Google Drive — You need a Google account and must authorize n8n to access your Drive. Create an OAuth2 credential in n8n for Google Drive.
2. Dumpling AI — Sign up at dumplingai.com and get an API key. Create a Header Auth credential in n8n with the key Authorization and value Bearer YOUR_API_KEY.
3. OpenAI — You need an OpenAI API key (from platform.openai.com). Create a generic API Key credential in n8n.
4. Airtable — You need an Airtable account and a base with a table. Get your Airtable API key from airtable.com/account. Create an API Key credential in n8n.

Steps to configure:

• Set up the Google Drive trigger to watch your specific folder.
• Update the Airtable nodes with your correct Base ID, Table ID, and column names.
• For the "Upload Blog post Image to Airtable" node, you must replace the URL template with your actual Airtable base ID and table ID (currently it shows placeholders). Also, ensure the JSON body uses the actual image URL from the Dumpling AI response, not the image prompt.
• Update the Dumpling AI API URLs if they change (check current documentation).

💡 Use Cases & Variations

• Content Repurposing: Automate turning webinar recordings, video lectures, or vlogs into blog posts with custom thumbnails.
• Social Media Content: Modify the workflow to post the blog and image directly to WordPress, Medium, or social platforms like LinkedIn and Twitter.
• Podcast to Blog: Use the same flow for audio files by adjusting the trigger to watch for audio files and the transcription endpoint.
• Multi-language Support: Change the GPT-4o prompt to generate blog posts in different languages.
• Enhanced Image Generation: Experiment with different Dumpling AI models (e.g., DALL-E, Stable Diffusion) by changing the model parameter in the image generation node.

Possible improvements: Add error handling (e.g., if transcription fails, send a Slack alert), or add a human review step before publishing. You could also use the image prompt to generate multiple image variations and select the best one.

{
  "id": "kyDSmFJkvi8CYHh2",
  "meta": {
    "instanceId": "a1ae5c8dc6c65e674f9c3947d083abcc749ef2546dff9f4ff01de4d6a36ebfe6",
    "templateCredsSetupCompleted": true
  },
  "name": "Auto-Generate Blog & AI Image from YouTube Videos with Dumpling AI & GPT-4o",
  "tags": [],
  "nodes": [
    {
      "id": "7847f8a8-a60e-4b2c-afa1-8d586cb42d5c",
      "name": "Watch Folder for New YouTube Videos",
      "type": "n8n-nodes-base.googleDriveTrigger",
      "position": [
        -2960,
        -420
      ],
      "parameters": {
        "event": "fileCreated",
        "options": {},
        "pollTimes": {
          "item": [
            {
              "mode": "everyMinute"
            }
          ]
        },
        "triggerOn": "specificFolder",
        "folderToWatch": {
          "__rl": true,
          "mode": "list",
          "value": "",
          "cachedResultUrl": "https://drive.google.com/drive/folders/1mde_V0ePcJEebVydygVKT7GDiABjj2A4",
          "cachedResultName": "Youtube Videos"
        }
      },
      "credentials": {
        "googleDriveOAuth2Api": {
          "id": "credential-id",
          "name": "googleDriveOAuth2Api Credential"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "cee77d1b-a91a-4fae-b5ae-eac34a9b0a03",
      "name": "Download Video File",
      "type": "n8n-nodes-base.googleDrive",
      "position": [
        -2740,
// ... truncated (copy to see full JSON)