Create viral social media videos with FalAI Flux/Kling and GPT-4 automation

AI-Powered Viral Video Factory 🚀

This workflow automates the entire process of creating short, cinematic, fact-based videos ready for social media. It takes a single concept, generates a script and visuals, creates video clips, adds a voiceover, and assembles a final video, which is then uploaded directly to your Google Drive. It's perfect for content creators and marketing agencies looking to scale video production with minimal manual effort.

How It Works 🎬

1. Generate a Viral Idea 💡: The workflow begins with the Create New Idea1 (OpenAI) node, which generates a viral-ready video concept, including a punchy title, hashtags, and a brief description based on a core theme (e.g., space, black holes). This idea is then logged in a Google Sheet.

2. Create a Cinematic Script & Voiceover 📜:

   • An OpenAI node (Generating scenes1) creates a detailed 12-scene script, outlining the visuals for a 60-second video.
   • The script text for all scenes is combined and prepared for voiceover generation by another OpenAI node (Generate Voiceover).

3. Generate Scene-by-Scene Visuals ✨: The workflow loops through each of the 12 scenes to create an animated clip:

   • Image Generation: An HTTP Request node sends the scene's prompt to the fal-ai/flux model to create a photorealistic still image.
   • Animation Prompting: The Video Prompts1 (OpenAI Vision) node analyzes the generated image and creates a new, specific prompt to animate it cinematically.
   • Image-to-Video: Another HTTP Request node uses the fal-ai/kling-video model to turn the still image into a 5-second animated video clip based on the new animation prompt.

4. Assemble the Final Video 🎞️:

   • Stitch Clips: Once all 12 clips are generated, the Merge Clips node uses the fal-ai/ffmpeg-api to concatenate them into a single, seamless 60-second video.
   • Add Audio: The Combine Voice and Video node then layers the AI-generated voiceover onto the stitched video.

5. Deliver to Google Drive 📂: Finally, the completed video is converted from a URL to a file and automatically uploaded to your specified Google Drive folder for easy access and publishing.

Key Technologies Used

• n8n: For orchestrating the entire automated workflow.
• OpenAI (GPT-4.1 & GPT-4o): For idea generation, scriptwriting, voiceover, and vision analysis.
• Fal.ai: For high-performance, API-based image generation (Flux), video animation (Kling), and video processing (FFMPEG API).
• Google Drive & Sheets: For logging ideas and storing the final video output.

Setup Instructions

1. Add Credentials:

   • In n8n, add your OpenAI API key.
   • Connect your Google account for Google Sheets and Google Drive access.
   • You will need a Fal.ai API Key.

2. Configure Fal.ai API Key:

   • Crucially, you must replace the placeholder API key in all HTTP Request nodes that call the fal.run URL.
   • Find the Authorization header in each of these nodes and replace the existing key with your own Key YOUR_FAL_AI_KEY_HERE.
   • Nodes to update: Create Images1, Get Images1, Create Video1, Get Video1, Merge Clips, Get Final video, Combine Voice and Video.

3. Configure OpenAI Nodes:

   • Select each OpenAI node (e.g., Create New Idea1, Generating scenes1) and choose your OpenAI credential.
   • You can customize the main prompt in the Create New Idea1 node to change the theme of the videos you want to generate.

4. Configure Google Sheets & Drive:

   • In the Organise idea, caption etc1 node, select your Google Sheets credential and specify the Spreadsheet and Sheet ID you want to use for logging ideas.
   • In the Upload file to drive node, select your Google Drive credential and choose the destination folder for your final videos.