Originally published at ffmpeg-micro.com
You can get ChatGPT to write scripts all day long. Hook ideas, product explainers, hot takes. It will generate dozens in minutes. But turning that text into an actual TikTok video? That's where most people stall out. They open CapCut or Premiere, spend 30 minutes per clip, and the "automation" dream dies on the editing timeline.
The bottleneck isn't the writing. It's the video production step. And you can eliminate it entirely with an API call.
The Pipeline: ChatGPT to TikTok in Four Steps
- ChatGPT generates a short-form video script via API.
- FFmpeg Micro burns that script onto a background video as a text overlay.
- n8n (or Make.com) orchestrates the whole thing and handles the download.
- The finished video gets pushed to TikTok through their content posting API or a scheduling tool.
No editing software. No manual exports. Once it's wired up, you trigger it on a schedule and videos show up in your TikTok drafts.
Step 1: Generate the Script with ChatGPT
In n8n, drop in an HTTP Request node pointed at the OpenAI chat completions endpoint. The key is your system prompt:
You write TikTok video scripts. Each script is under 60 seconds when read aloud.
Use short punchy sentences. One idea per script. No intros, no outros.
Output ONLY the script text, nothing else.
Keep scripts under 100 words. TikTok viewers scroll fast. Aim for 40 to 70 words per script.
Step 2: Create the Video with FFmpeg Micro
FFmpeg Micro has a virtual option called @text-overlay that burns text directly onto a video file through a single API call:
curl -X POST https://api.ffmpeg-micro.com/v1/transcodes \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"inputs": [{"url": "https://your-bucket.s3.amazonaws.com/background-clips/clip-01.mp4"}],
"outputFormat": "mp4",
"options": [
{
"option": "@text-overlay",
"argument": {
"text": "Your ChatGPT-generated script goes here",
"style": {
"charsPerLine": 20,
"fontSize": 48,
"lineSpacing": 12,
"x": "(w-text_w)/2",
"y": "(h-text_h)/2",
"boxBorderW": 10
}
}
}
]
}'
The @text-overlay option handles font sizing, centering, and line wrapping for you.
Putting It Together in n8n
The workflow is four nodes in a line:
- Schedule Trigger fires on your cadence.
- HTTP Request (ChatGPT) generates the script.
- HTTP Request (FFmpeg Micro) creates the video with text overlay.
- HTTP Request (Download + Post) grabs the output and pushes it downstream.
If you prefer Make.com over n8n, the same logic applies.
Why This Works for Marketing
Each FFmpeg Micro transcode costs fractions of a cent. ChatGPT API calls for short scripts run well under a penny. You could produce 100 videos a week for less than the cost of a coffee. You set it up once and the workflow runs on a schedule.
FAQ
Can I use ChatGPT to create TikTok videos automatically?
Yes. Connect the ChatGPT API to FFmpeg Micro using n8n or Make.com. ChatGPT generates the script, FFmpeg Micro burns it onto a video, and the result uploads to TikTok without manual editing.
What does it cost to automate TikTok video creation with AI?
Each video costs less than one cent to produce. Running 100 videos per week costs a few dollars total.
Do I need to know FFmpeg to use FFmpeg Micro for TikTok videos?
No. FFmpeg Micro's @text-overlay option handles text rendering through a JSON API. You don't write FFmpeg command-line filters.
Does TikTok allow automated video uploads?
TikTok offers a Content Posting API for developers. You need to apply for access through their developer portal.
Top comments (0)