A 55 second storytime Short about the legend of how coffee was discovered. Voice, paper-cut scenes, café footage and word-by-word captions, all made by Claude in Editbay. Nobody filmed anything.
About this film
The goats that discovered coffee
Format
9:16 for YouTube Shorts, TikTok and Reels
Length
55 seconds, 25 scenes
Voice
One continuous AI voiceover take (ElevenLabs)
Pictures
Paper-cut scenes built in HTML, plus café footage from Mixkit
A faceless video tells a story without anyone on camera. A voice carries it, and the pictures change every two or three seconds: b-roll, illustrations, simple animation, with bold captions so it works on mute. It is the format behind most faceless YouTube channels, and the one people search for most when they want to make videos without showing themselves.
It works because the story does the heavy lifting. A question in the first two seconds, a few quick beats, a twist, and a line that invites a comment. The film here follows that pattern exactly, so you can copy the structure for any topic: history, science, true stories, product origins, explainers.
The structure
The beats this film follows. Copy them for your own topic.
The hook (0 to 3 s)Open a question the viewer needs answered. Here: "Coffee was discovered by goats. Supposedly."
The setupWho, where and what was strange, in two or three short lines. A new picture on every line.
The story beatsFour or five beats, each with one clear image: the dancing goats, the herder tasting the berries, the monk and the fire, the first brew.
The twistTurn the story once. Here: nobody wrote it down until 1671, so it is probably only a legend.
What we do knowOne or two true facts that ground the story, so the video teaches something real.
The comment lineEnd on a question people want to answer: "Do you believe the goats?"
How to make a faceless video with your AI
Editbay is the video studio your own AI works in. Claude or ChatGPT does the creative work; Editbay gives it the editing server, the asset library, rendering and checks. This is how this film was made.
Connect your AIAdd Editbay as a connector in Claude or ChatGPT. Your AI gets the editing server, the asset library and the render engine.
Ask for the videoDescribe the story, the length and the look in a sentence or two. The prompt below is what made this film.
Settle the choices firstYour AI researches the facts, then puts the decisions in front of you: voice, pace, look and music, each with a real sample. You pick; nothing is built on a guess.
Script and one voice takeThe whole script is recorded as one take, so the delivery flows. Editbay transcribes it and cuts it into scenes at the word timings.
Scenes and captionsYour AI builds one scene per line (illustrations in HTML, stock clips where they fit) and burns in word-by-word captions inside the safe zone.
Review with notes on the frameWatch it, pin notes to the exact moment ("bigger captions here"), and your AI makes the next version from your notes.
The ask behind this film
Make a 50 second faceless storytime Short about the legend of how coffee was discovered. Paper-cut illustrations for the legend, café footage for today, a warm storyteller voice and bold word-by-word captions. Tell it as a legend, not as history, and end with a question for the comments.
Tips for a faceless video that works
Write for the ear. Short sentences, one idea each. Every sentence becomes a scene.
Change the picture every two to three seconds. Faceless videos lose viewers on static frames.
Captions are half the video. Most people watch on mute, so make them big and keep them inside the safe zone.
Record the voiceover as one take. Line by line sounds stitched together.
Keep facts true. Say "legend says" when it is a legend.
The script
Coffee was discovered by goats. Supposedly. Legend says an Ethiopian goat herder named Kaldi noticed something strange. His goats nibbled little red berries. And then... they started dancing. All night long. So Kaldi tried the berries himself. Now he was dancing too. He took them to a monastery. A monk frowned, and threw them in the fire. But then... that smell. They raked out the beans, crushed them, and stirred them into hot water. The first coffee. Great story. One problem: nobody wrote it down until sixteen seventy-one. Centuries later. What we do know: coffee's roots are in the Ethiopian highlands. In the fourteen hundreds, Sufis in Yemen drank it to stay awake at night. So the dancing goats? Probably just a legend. Do you believe the goats? Tell me in the comments.
Questions
What is a faceless video?
A video where nobody appears on camera. A voiceover tells the story over b-roll, illustrations or animation, usually with big captions. Faceless YouTube channels and TikTok accounts are built on this format.
Can AI make a whole faceless video?
Yes. This one was made entirely by Claude in Editbay: the research, the script, the voice, every scene, the captions and the edit. The only human input was the brief and the choices.
How long should a faceless Short be?
Long enough to tell one story well, usually 30 to 60 seconds. Cut every line that does not move the story forward.
Do I need my own footage?
No. Illustrated scenes are generated as code, stock clips come from free libraries, and with Premium your AI can generate video clips and images too.
Credits
Sound: "Fire - Effects - Swooshes - Bursts, evolving" by GregorQuendel (freesound.org/s/421879), CC BY 4.0.
Sound: "Fire Torch_whoosh 1 (medium speed).wav" by kev_durr (freesound.org/s/426208), CC BY 4.0.
Sound: "Single Goat Bleating 2x" by Kinoton (freesound.org/s/581240), CC0.
