Automate Faceless AI Videos for TikTok and Reels
Faceless videos are everywhere on TikTok and Instagram Reels. An AI-generated voice narrates over AI-generated images, with captions burned in. No camera. No face. No editing software. Some of these accounts pull millions of views per month.
Building this pipeline from scratch requires stitching together four or five tools: one for script writing, one for image generation, one for voiceover, one for video editing, and one for publishing. On the n8n forum, threads about automating short-form video creation consistently rank among the most active. On Make.com, users share elaborate scenarios with 15+ modules that break when any provider changes their API.
Here is a three-node workflow that handles the entire pipeline.
What a faceless video pipeline looks like
A faceless video has four components:
A script (what the narrator says)
AI-generated images (the visuals behind the narration)
An AI voiceover (text-to-speech reading the script)
Captions (text on screen matching the voiceover)
Without Blotato, you need a separate API for each: OpenAI for the script, Midjourney or DALL-E for images, ElevenLabs for voice, and FFmpeg or Remotion for stitching it together. Each API has its own authentication, rate limits, and failure modes.
With Blotato's Create Visual endpoint, you send a prompt. Blotato generates the script, images, voiceover, and captions, then renders the final video. One API call.
The workflow: three steps
Step 1: Generate the video
Send a request to the Create Visual endpoint with the "AI Video with AI Voice" template:
{
"templateId": "5903fe43-514d-40ee-a060-0d6628c5f8fd",
"inputs": {},
"prompt": "Create a 60-second video about 5 productivity habits of successful CEOs. Use a calm, authoritative male voice."
}The prompt field tells the AI what to create. Blotato fills in all template inputs automatically: script, image descriptions, voice selection, and caption styling. The response returns a video ID:
Step 2: Wait for rendering
Poll the Get Visual Status endpoint with the video ID. Rendering takes 1-3 minutes for a 60-second video. When the status reaches done, the response includes mediaUrl with your rendered video file.
Step 3: Publish to TikTok and Reels
Pass the mediaUrl from Step 2 to the Publish Post endpoint.
TikTok:
Instagram Reels:
YouTube Shorts:
Step-by-step: n8n workflow
Install the Blotato community node in n8n.
Add a trigger node. Use a Schedule node to create one video per day, or a webhook to trigger on demand.
Add an OpenAI node (or any LLM) to generate your video topic and prompt. Feed trending topics, RSS feeds, or a list of pre-written topics. This step is optional -- you write the prompt directly in the Blotato node if you prefer.
Add a Blotato Create Visual node. Select the "AI Video with AI Voice" template. Pass your prompt.
Add a Blotato Get Visual Status node. Enable polling until status is
done.Add Blotato Publish Post nodes in parallel -- one for TikTok, one for Instagram, one for YouTube. Pass the
mediaUrlfrom the Get Visual Status node into each node'smediaUrlsfield.Activate the workflow.
For a full video walkthrough and importable template, see: n8n Faceless Videos
Step-by-step: Make.com scenario
Add a trigger module (schedule, webhook, or Google Sheets).
Add an HTTP module to call
POST https://backend.blotato.com/v2/videos/from-templateswith your template ID and prompt.Add a polling loop (HTTP module + Sleep module) to check
GET https://backend.blotato.com/v2/videos/{id}until status isdone.Add a Router module.
Add one HTTP module per platform, each calling
POST https://backend.blotato.com/v2/postswith the rendered video URL.Activate the scenario.
For a full walkthrough, see: Make Faceless Videos
Available video templates
The "AI Video with AI Voice" template is the most popular for faceless content, but other templates work too:
Narrated faceless videos with AI images
AI avatar reading your script
AI presenter with generated background footage
Slideshow-style video from your own images
Browse all templates at my.blotato.com/videos/new or query the API: GET /v2/videos/templates?fields=id,name,description,inputs
Scaling to one video per day
Pair this workflow with Blotato's Source API to generate topics automatically:
Use a
perplexity-querysource to research trending topics in your niche daily.Feed the research output into your LLM node to write a video prompt.
Pass the prompt into the Create Visual node.
Publish to all platforms.
One workflow. One trigger. One video per day across three platforms with zero manual work.
Forum threads referenced
These are the community discussions that prompted this post:
Last updated