Generate 24-second ‘what if’ videos with Gemini 2.5 Pro and Google Veo 3.0

How it works

1. Generate the script – Takes a "what if" question and creates a 24-second video script with 4 scenes (Hook, Wonder, Reality, CTA) using Gemini 2.5 Pro

2. Create Veo prompts – Translates each scene's visual description into a production-ready prompt (120–150 words) optimized for Google Veo 3.0

3. Generate voiceover – Converts each scene's script into natural speech using Google Text-to-Speech and uploads to GCS

4. Generate video clips – Creates four 8-second AI-generated video clips with Veo 3.0 and uploads each to GCS

5. Merge & render – Combines all video clips and audio tracks using Creatomate into a single 24-second video

6. Deliver – Sends the final video back via Telegram

Set up steps

1. API Keys – Set up accounts for Google Cloud (Veo 3.0 + Text-to-Speech + GCS), DeepSeek, Creatomate, and Telegram Bot

2. Credentials in n8n – Add each API key as a credential in your n8n instance

3. Configure variables – Update the SET node with your Google Cloud PROJECT_ID and CLIENT_EMAIL

4. Test – Enter a "what if" question and execute the workflow