Originally published at ffmpeg-micro.com
Faceless TikTok channels post 3-5 times a day. No filming, no editing software, no face on camera. The best ones gross $2K-10K/month from the Creator Fund and brand deals. And the entire content pipeline can be automated.
The bottleneck isn't idea generation or scripting. It's video assembly. Taking a script, matching it with stock footage, adding TTS narration, burning in captions, and exporting a vertical video takes 20-30 minutes per clip when done manually. At 5 clips a day, that's your entire workday gone.
The fix: an API-driven pipeline. You write the script (or let an LLM do it), generate narration, pull stock footage, and send everything to FFmpeg Micro to compose the final video. The whole thing runs in the background.
The pipeline
Every faceless video follows the same five steps:
- Script from a prompt template (ChatGPT, Claude, or any LLM)
- TTS audio from the script (ElevenLabs, OpenAI TTS, or PlayHT)
- Stock footage matching the topic (Pexels API, Pixabay, or your own library)
- Video composition layering footage, audio, and captions (FFmpeg Micro)
- Upload to TikTok (scheduled via their API or a tool like Publer)
Steps 1-3 produce raw inputs. Step 4 is where FFmpeg Micro does the work.
The composition step
This is where most people get stuck. You have a background video, an audio track, and caption text. You need a 9:16 vertical video with everything synced and styled.
With FFmpeg Micro, that's one HTTP request to POST https://api.ffmpeg-micro.com/v1/transcodes:
{
"inputs": [
{ "url": "https://example.com/stock-footage.mp4" },
{ "url": "https://example.com/tts-narration.mp3" }
],
"outputFormat": "mp4",
"options": [
{ "option": "-c:v", "argument": "libx264" },
{ "option": "-c:a", "argument": "aac" },
{ "option": "-shortest", "argument": "" }
]
}
FFmpeg Micro overlays the audio onto the video, trims to the shorter track's duration, and returns a composed MP4. The -shortest flag ensures the output matches whichever input ends first.
Want captions burned into the video? Add a text overlay:
{
"inputs": [
{ "url": "https://example.com/composed-video.mp4" }
],
"outputFormat": "mp4",
"options": [
{
"option": "@text-overlay",
"argument": {
"text": "Your caption text goes here for maximum viewer engagement",
"style": {
"charsPerLine": 18,
"fontSize": 60,
"lineSpacing": 15,
"x": "(w-text_w)/2",
"y": "0.75*h",
"boxBorderW": 12
}
}
}
]
}
The @text-overlay virtual option handles word-wrap, positioning, and styling automatically. No FFmpeg filter graph knowledge required.
Automating the full pipeline with n8n
The real leverage comes from connecting all five steps. n8n is the most popular choice for this kind of workflow (and it's free to self-host).
The flow looks like this:
- Trigger: Schedule node fires every 4 hours
- Script: HTTP Request to OpenAI API with your niche-specific prompt template
- TTS: HTTP Request to ElevenLabs with the script text, returns an audio URL
- Footage: HTTP Request to Pexels API, searching keywords extracted from the script
- Compose: HTTP Request to FFmpeg Micro combining footage + audio + overlay text
- Post: HTTP Request to TikTok's API (or Publer) to schedule the upload
Each node passes its output to the next. The workflow runs unattended. You wake up to fresh TikTok posts every morning.
For the FFmpeg Micro step, use n8n's HTTP Request node. Point it at https://api.ffmpeg-micro.com/v1/transcodes, set the auth header to Bearer YOUR_API_KEY, send the JSON body, and poll the returned job ID until the status field shows completed.
What this actually costs
Running a 5-post-per-day faceless channel through this pipeline:
| Component | Cost | Notes |
|---|---|---|
| LLM scripting | ~$1.50/month | GPT-4o-mini at $0.01/script |
| TTS narration | ~$15/month | ElevenLabs at ~$0.15/minute for 60s clips |
| Stock footage | Free | Pexels API, no cost |
| FFmpeg Micro | $19/month | Pro plan, up to 100 minutes |
| Total | Under $40/month | For 150 videos |
Compare that to hiring an editor ($500+/month) or spending 2.5 hours daily doing it yourself.
Three formats that work
Not every niche fits this pipeline. The best faceless formats share one trait: the visuals don't need to be original. Stock footage or simple graphics carry the story while narration does the real work.
Motivational/quote channels pair atmospheric footage with punchy narration. High volume, lower per-video engagement, but the algorithm rewards consistency.
Explainer channels (psychology facts, history, tech tips) use relevant B-roll with informational TTS. These build loyal audiences and monetize well through brand deals.
News recap channels combine stock footage with scripted summaries of trending topics. High discoverability, but they need fresh scripts daily.
All three use the same pipeline. The only things that change are the prompt template and the footage search queries.
Common gotchas
Aspect ratio matters more than resolution. TikTok wants 9:16. If your stock footage is 16:9, you need to crop or pad it. FFmpeg Micro handles this through crop and scale options in the request body.
TTS pacing affects watch time. Too fast and viewers swipe away. ElevenLabs lets you control speed. Target 140-160 words per minute for TikTok.
Captions aren't optional. 85% of TikTok is watched on mute. Burned-in captions using the @text-overlay option aren't just for accessibility. They're a retention tool.
Batch your jobs. Instead of composing one video at a time, queue 5-10 transcode requests in parallel. FFmpeg Micro processes them concurrently and returns results independently.
Get started
Sign up for a free FFmpeg Micro API key at ffmpeg-micro.com. Upload a stock clip, overlay some TTS audio, and see the composed result in under a minute.
For the full n8n walkthrough with a working template, check out the Viral Shorts Automation training.
FAQ
Can I really run a TikTok channel without showing my face?
Yes. Faceless channels are one of the fastest-growing categories on TikTok. Channels like motivation compilations and fact explainers get millions of views using only stock footage and TTS narration. No filming required.
How many videos should I post per day?
Start with 2-3 and scale based on performance. TikTok's algorithm rewards consistency over production quality. With this pipeline, scaling from 3 to 5 posts per day means changing a single cron schedule.
Will TikTok flag AI-generated content?
TikTok requires disclosure of AI-generated content in some regions. Add a note in your bio or video description. The platform doesn't currently penalize AI content algorithmically, but policies evolve.
Do I need to self-host n8n for this?
No. n8n Cloud starts at $20/month. But self-hosting on a $5/month VPS gives you unlimited workflow executions and keeps the total pipeline cost lower.
How much can a faceless TikTok channel actually earn?
The TikTok Creator Fund pays roughly $0.02-0.04 per 1,000 views. A consistent channel hitting 100K views/day (realistic after 3-6 months of daily posting) earns $2-4/day from the fund alone. Brand deals, affiliate links, and cross-posting to YouTube Shorts and Instagram Reels can multiply that 5-10x.
