Automate Tiktok Slideshows with Music (n8n and Make.com)
This tutorial demonstrates how to automate TikTok slideshows with trending music using n8n and Make.com. The workflow uses a reference image for visual consistency, generates AI image prompts via OpenAI, creates images through Replicate's Nano Banana model, and publishes to TikTok via Blotato. Both platforms are covered with step-by-step API setup instructions.
Summary
The video is a walkthrough tutorial hosted by Ding and Kevin, an AI automation expert from Malta, showing how to automate TikTok photo slideshows with auto-applied trending music. The core use case demonstrated is a pet-related product promotion using a consistent reference dog image across all slides, though the hosts emphasize the workflow is easily adaptable to any brand or product with UGC imagery.
The Make.com portion is covered first as the simpler entry point. Users configure an 'edit here' module with three key inputs: a reference UGC image URL (publicly accessible, e.g., from Google Drive), a topic description, and a desired number of slides. The system automatically prepends the original reference image to the slideshow, so requesting 3 slides yields 4 total images. An OpenAI module is used to generate structured image prompts for each slide, including captions, titles, and a hook for the first slide, all based on a pre-built system prompt users don't need to modify.
Replicate is used as the image generation service, specifically the Nano Banana model, chosen because it accepts a reference input image for visual consistency. Replicate is recommended for its pay-per-usage pricing model with no monthly subscription, making it cost-effective for variable workloads. The Replicate API token is obtained from the platform's API tokens section and inserted into Make.com using the format 'Bearer [space] API token.'
Blotato serves as the TikTok publishing layer, handling media URL mapping, captions, slideshow titles, and critically, the auto-add music feature that automatically selects trending TikTok music. Users can also choose to post as drafts to manually select trending sounds, or use Blotato's scheduling feature with 'next free slots' to maintain a consistent posting calendar (e.g., Monday–Friday at 10 AM).
The N8N section mirrors the Make.com workflow structurally, using a sticky note editor interface for the same three inputs. The same OpenAI system prompt, Replicate integration, and Blotato publishing setup are used, with the key difference being the N8N-specific credential configuration. A common error noted is accidentally removing the space between 'Bearer' and the API token. Both platforms produce the same end result: an automated, music-enhanced TikTok slideshow posted on a defined schedule.
Key Insights
- Kevin explains that the reference image feature is specifically valuable for brands with UGC assets, as it ensures visual consistency of a specific product or image across all generated slideshow frames.
- The hosts clarify that requesting 3 slides results in 4 total images in the final slideshow, because the original UGC reference image is automatically prepended to the generated slides.
- Replicate is recommended over subscription-based image generation services specifically because it charges per usage with no monthly fee, meaning users pay $0 in months they generate nothing.
- Blotato's auto-add music feature automatically selects trending TikTok music for slideshows, but users can alternatively post as drafts to manually choose a specific trending sound within TikTok itself.
- A commonly noted error in both Make.com and N8N setups is accidentally removing the space between 'Bearer' and the Replicate API token in the authorization header, which breaks the connection.
Topics
Full transcript available for MurmurCast members
Sign Up to Access