Back to Tutorials
Intermediate20 minutes
AI-Powered Video Generation
Create automated video content using AI nodes for script generation, voice synthesis, and image generation.
What You'll Build
An automated workflow that generates complete videos from a simple text prompt, including AI-generated script, voiceover, and visuals.
- Generate video scripts using ChatGPT or Claude
- Create images with Stable Diffusion
- Add professional voiceovers with ElevenLabs
- Combine everything into a final video
Step 1: Set Up Your Prompt
Start with an On Play trigger and a String Input node for your video topic.
Example prompt:
"Create a 30-second video about space exploration"Step 2: Generate Script with AI
Add a ChatGPT or Claude node to generate your video script.
Node Configuration:
- Model: GPT-4 or Claude Sonnet
- System Prompt: "You are a video script writer"
- User Prompt: Connect to your String Input
- Temperature: 0.7 for creative but focused output
Step 3: Generate Visuals
Use Stable Diffusion to create images based on your script.
Image Generation Settings:
- Extract key scenes from the AI-generated script
- Generate 3-5 images for different video segments
- Resolution: 1920x1080 for HD video
- Style: "cinematic, high quality, professional"
Step 4: Add Voiceover
Connect an ElevenLabs node to convert your script to natural-sounding speech.
Voice Settings:
- Voice: Choose a professional narrator voice
- Stability: 0.5 for natural variation
- Clarity: 0.75 for clear pronunciation
- Input: Connect to the script output from ChatGPT
Step 5: Assemble the Video
Use Video Operations to combine your generated images and add transitions.
- Add Video Transitions node for smooth scene changes
- Add Audio Track node to sync your voiceover
- Use Video Rendering to export the final video
Advanced Tips
- Use Variables to store and reuse AI outputs across your workflow
- Add Stock Video nodes for B-roll footage to enhance your generated content
- Create a webhook trigger to automate video generation from external systems