Generate addictive ASMR Veo 3 reels

Go to Workflow
0 views
Built by Koulikas Giannis Koulikas Giannis
Created on June 07, 2026

Description

How it works

1. Generate new idea – Pulls previously used objects from Google Sheet, then uses GPT-4.1-mini to generate a new fruit-shaped glass object not in the list

2. Create Veo prompt – Uses GPT-4.1-mini to generate a hyper-realistic, ASMR-style prompt for the object being gently divided (with safety guidelines to avoid policy triggers)

3. Generate video – Creates an 8-second AI video with Veo 3.0 showing the glass fruit being elegantly separated on a cutting board with ASMR sound layers

4. Convert to vertical – Crops the 16:9 video to 9:16 vertical format using Shotstack

5. Upload to Postiz – Uploads the final video to Postiz (social media scheduler)

6. Schedule across platforms – Detects connected channels (TikTok, YouTube, Instagram) and schedules the video for posting

7. Log for future – The new object is logged to Google Sheets to avoid repetition in the next run

Set up steps

API Keys – Set up accounts for Google Cloud (Veo 3.0 + GCS), OpenAI, Shotstack, and Postiz

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

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

Set up Google Sheet – Create a sheet with a column named object to track previously used objects

Connect Postiz – Ensure your social media channels (TikTok, YouTube, Instagram) are connected in Postiz

Test – Execute the workflow manually or let the schedule trigger run every 8 hours

Nodes Used (7)

AI Agent
@n8n/n8n-nodes-langchain.agent
Google Cloud Storage
n8n-nodes-base.googleCloudStorage
Google Sheets
n8n-nodes-base.googleSheets
HTTP Request
n8n-nodes-base.httpRequest
JWT
n8n-nodes-base.jwt
OpenAI Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenAi
Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured