Create voiced real estate tour videos from CSV listings with Claude and Shotstack
Quick overview Put your listings in a CSV in Google Drive and get a narrated tour video for each. Claude looks at every photo, writes one line per room, and each photo stays on screen exactly while the voice talks about it. Captions, price, key facts and an agent end card included.
How it works Downloads Shotstack's guide for AI agents, reads your listings CSV from Google Drive and works through it one listing at a time. Check the listing stops a row with no address, no price or fewer than 3 photo links before anything costs money. Look at the photos sends every photo to Claude Opus 5.5, which notes the room each one shows and what a buyer would notice in it, and flags repeat angles and virtually staged photos. Claude makes the video picks the best 8 to 12 photos for a tour, outside first and the view last, and writes one voice line per photo, using only the facts in your row. Each line is its own voice clip, and each photo takes its timing from its line, so the picture changes exactly when the voice moves to the next room. Captions, the price, the facts, room labels and the end card follow the same timing. Check the edit makes sure every photo is timed to its line and the price is shown exactly. Claude fixes anything else, up to twice per listing. Shotstack renders the MP4 and Save the video puts it in Google Drive.
Setup Add the Shotstack node from the nodes panel. It's a verified node, so it works on n8n Cloud and self-hosted. Then import this workflow. Make a CSV with the columns address, price, beds, baths, size, highlights, description, photos, agent, contact, logo and voiceover. Put the photo links in one cell, separated by spaces, paste the listing text into description, and write yes or no under voiceover. Upload the CSV to Google Drive and pick it in Download the listings. Pick a folder for the videos in Save the video, and connect Google Drive in both steps. Add your Anthropic API key to Look at the photos, Claude and Claude (fixes), and your Shotstack API key to Make the video, Wait for the video and Get the video link. Free trial: https://shotstack.io/?utm_source=n8n&utm_medium=template&utm_campaign=listing-reels&utm_content=description Then click Execute workflow.
Requirements n8n Cloud or self-hosted, with the Shotstack node, version 0.2.2 or later A Google Drive account for the CSV and the videos A Shotstack API key. An Anthropic API key with Claude Opus 5.5, or another Claude model that can read images
Customization Change the shape in Video size: 1920 x 1080 landscape to start with, 1080 x 1920 for Reels, TikTok and Shorts, 1080 x 1080 square or 1080 x 1350 for feed posts Write no under voiceover for a music-only video with the same tour Add a logo link to a row to put the agency logo on the end card Edit The brief to change the tour order, the length, the fonts or the colours Read your listings from Google Sheets or Airtable instead of the CSV
Additional info Only the listing's real photos are used: nothing is AI-generated, and staged furniture is never described as included. Photos 1920 pixels wide or larger look sharpest in landscape. Each video takes about 10 to 15 minutes, because Claude thinks at extra-high effort, and costs about 1.5 Shotstack credits plus your Claude usage.
Related Templates
Track Demo Bookings with Google Calendar to Meta Conversions API Integration
Who is this workflow for? If you're using Meta Ads to generate new leads to your sales pipeline, this workflow is for yo...
Build an MCP Server which answers questions with Retrieval Augmented Generation
Build an MCP Server which has access to a semantic database to perform Retrieval Augmented Generation (RAG) Tutorial C...
Transcribe & Summarize Audio with Whisper and GPT, from Google Drive to Notion
This workflow contains community nodes that are only compatible with the self-hosted version of n8n. Overview This work...
🔒 Please log in to import templates to n8n and favorite templates
Workflow Visualization
Loading...
Preparing workflow renderer
Comments (0)
Login to post comments