n8n template · self-hosted
Faceless Video Machine
Give it a topic and get a finished, animated vertical video on your phone: script, natural voice, motion-design scenes in your brand colours and captions that pop. You tap “Ready to post”. It costs $0 per video.

Not stock footage glued together: every scene is designed and animated to the voice. A hook in big letters, lists that build up card by card, steps on a timeline, before/after cards, a chat on a phone, your end card. And once per video, a short clip of real people for a human touch.
Who it’s for
Faceless channel creators, social media managers, small marketing agencies and small businesses (a coffee shop, an online shop, an estate agent) that want a short video every day without filming.
What it does
- Writes the script scene by scene with free AI (Gemini), with a different hook style every time so your videos don’t look mass-produced.
- Records the voice-over with free Gemini TTS, in English or Spanish, in the voice you choose. Four free voice models in a row: if one is busy, the next one steps in.
- Animates it like a studio: 8 scene types we designed, synced to the voice, in your brand colours. Word-by-word captions with the key word in colour, a progress bar, your @, soft music and sound effects we made ourselves. 1080×1920 MP4.
- Adds a human touch: one short clip of real people from Pixabay (free licence), never presented as a customer or a testimonial.
- Asks you first: the video arrives on Telegram. Tap Ready to post and you get the title, caption and hashtags to paste. Nothing is posted on its own.
- On autopilot: one new video every day at 10:00, or on demand from a simple form.
Measured, not promised
- We ran it 10 times in a row on our own n8n, in English and Spanish: 10 of 10 videos came out complete (1080×1920, voice, word-by-word captions, animated scenes). Median time from topic to finished MP4: 76.4 s (script 3.4 s, voice 19.8 s, animation 51 s on our 8-core computer; a slower one takes longer; voice measured with two stand-in Gemini voice models because the first one was out of quota that day). Average length 37.5 s; biggest file 29.5 MB, under Telegram’s 50 MB. In 1 of them a voice model failed and a backup model finished the job. Cost per video: $0. (3 October 2026)
What you get
- The ready-to-import n8n workflow, in English and Spanish.
- Our motion-design video renderer, with its Docker setup: one command (
docker compose up) starts n8n and the renderer together. - A step-by-step PDF guide (5 steps to your first video).
What you need
Docker on your computer or server (n8n comes included), a free Gemini key and a free Telegram bot. Optional: a free Pixabay key for the clip of real people. It does not run on n8n Cloud: the renderer runs next to your own n8n.
Faceless Video Machine
$29 one-time
30-day guarantee, no questions asked: if it isn’t right for you, for any reason, you get your money back. How refunds work.
Get it on GumroadYou may also like
Scripts + n8n templateFaceless Channel Voice StudioA queue of topics becomes finished voice-overs every morning: checked script, directed AI voice, your OK on Telegram.$29
n8n templateContent MachineOne idea, typed or by voice, becomes LinkedIn, Instagram and X posts in your voice, plus 7 ideas every Monday.$29
Free kit · 4 automationsFree AI kit for TelegramSummarise any link, turn voice notes into text, get a morning news digest and an alert when a website changes.Free← All templates · amelqa.com home
For use in your business. Reselling the files is not allowed. The renderer uses Remotion, free for individuals and companies of up to 3 people (bigger companies need a Remotion licence). Clips under the Pixabay Content License. Gemini, Pixabay, Telegram, TikTok, Instagram, YouTube, Docker, Remotion and n8n are trademarks of their owners; no affiliation with Amelqa.