To turn a text into an animated video, don't paste the whole text into a tool: pull one question out of it, rewrite the answer for the ear (about 150 to 180 words per minute in English), then split it into scenes, one sentence per scene, each with a picture that shows what the voice says. A generator like ClipOnIt's Explainer handles the scene split, the animated illustrations, the voice and the captions; the human work is rewriting and reviewing. The example in this article, "Why onions make you cry", fits in 84 words, 7 scenes and 26 seconds.
Which texts can become an animated video?
Almost any text that explains something: a blog post, a product sheet, lesson notes, a LinkedIn post, an FAQ, a script you already wrote. The real obstacle is length. A few benchmarks:
| Starting text | Typical length | What it becomes |
|---|---|---|
| LinkedIn post or paragraph | 80 to 150 words | One 30 to 60 s video, almost as is |
| Product sheet or FAQ | 200 to 400 words | 2 to 4 videos, one per question |
| Blog post | 1,000 to 2,000 words | 5 to 10 videos, one per subheading |
| Course chapter | 3,000 words or more | A series; each concept becomes a video |
These assume the pace we measure on our own English videos, around 170 to 190 words per minute. A 1,500-word article read in full would make an 8-minute video: nobody would watch it on a phone.
The method in 5 steps
1. Boil the text down to one question
A short video answers one question. Reread your text and write the question it answers best, in one line, the way someone would type it: "Why do onions make you cry?", "How do I cancel a subscription online?", "What does our new feature do?". The rest of the text can wait for other videos.
2. Rewrite for the ear
Text written to be read sounds stiff out loud. For speech: sentences of 8 to 15 words, the subject first, no parentheses, simple numbers. The answer comes in the first sentence; the explanation follows. This pass often halves the length.
3. Split into scenes
A scene is one sentence, sometimes two: 3 to 8 seconds on screen. That's the pace that holds attention on TikTok, Reels and Shorts; a picture that stays more than 8 seconds loses people. A 30-second video usually has 6 to 9 scenes.
4. One picture per sentence, timed to the key word
Every sentence has one word that carries the idea ("gas", "cells", "tears"). That's what the picture must show, at the exact moment the voice says it. It's the difference between an explainer video and a slideshow with a voice-over.
5. Voice, captions and render
A synthetic voice or your own, word-by-word captions (a large share of viewers watch with the sound off), then a vertical 9:16 render for social media.
Example: 7 sentences, 7 scenes, 26 seconds
Here is a real video produced by the Explainer from a 7-sentence script, in the Cutaway house style (a 3D dollhouse, room by room). Voice-over: Mathis, ClipOnIt's founder.
The text and how it was split:
| Scene | Sentence read | On-screen title |
|---|---|---|
| 1 | You're not crying because of the onion. You're crying because of a gas. | It's not the onion, it's a gas |
| 2 | An onion is thousands of tiny cells. When the knife breaks them, they release a gas that rises to your eyes. | Broken cells release a gas |
| 3 | Your eyes fight back: they make tears to rinse it away. | The eyes fight back with tears |
| 4 | Three tricks. One: a sharp knife crushes fewer cells. | A sharp knife |
| 5 | Two: a cold onion, straight from the fridge, releases less gas. | A cold onion |
| 6 | Three: cut near an open window or under the hood. | Open window or hood |
| 7 | And if you still cry, just say it's emotional. | Just say it's emotional |
What to take from it for your own text:
- The answer is in scene 1. Someone who drops off after 3 seconds still learned something.
- Each on-screen title is 3 to 6 words. It sums up the sentence; it doesn't repeat it.
- The text moves from explanation to practice (scenes 4 to 6): that's what makes people save the video.
- The last line is a punchline: it ends on the topic, with no call to subscribe.
Read as is, or rewritten by AI?
In ClipOnIt's Explainer, there are three starting points:
- A finished text: it's read as is and split into scenes. Choose this when the text is already written for speech.
- An idea or a topic: AI writes the narration from your key facts, then splits it. Handy for starting from a long article: give the question and the three facts that matter (see from one sentence to an animated video).
- Your voice: you read the text on the mic or import audio; the narration is transcribed and each element appears on your words.
Every scene then stays editable: ask the AI Director for a change ("shorter", "a chart here"), compare 3 variants of a scene with the same text, or move an element by hand. The final render is a vertical MP4.
What AI does, and what's left for you
| Step | Done by the tool | Up to you |
|---|---|---|
| Choosing the question | No | Yes: it's the most important decision |
| Rewriting for speech | A first draft | Read it out loud, cut |
| Splitting into scenes | Yes | Merge or split a scene if needed |
| Illustrating and animating | Yes, from a library of animated elements | Replace a picture that shows the wrong thing |
| Voice and captions | Yes | Listen to names and numbers |
| Fact-checking | No | Yes, every figure |
Mistakes that make the video flat
- Pasting the whole article. The video runs too long and answers nothing specific.
- Keeping written sentences. A 30-word sentence with two commas becomes a 10-second scene on a single picture.
- Decorative pictures. If the picture doesn't show what the voice says, it distracts.
- Changing style every video. For a series, keep the same visual world: it's your signature. Choosing a style is covered in which explainer video style to pick.
FAQ
Can I turn a PDF or an article into a video directly?
In the Explainer, the starting point is a pasted text, an idea or an audio recording. For a PDF or an article, copy the part that answers your question, or give the question and the key facts: the video will be better than with the whole document.
How long does it take to turn a text into an animated video?
Rewriting takes 5 to 15 minutes for a 30 to 60-second video. Scene generation is automatic, and rendering takes about 4 minutes on ClipOnIt. These are orders of magnitude, not a measurement.
Do I need animation skills?
No. The elements are already drawn and animated; the tool picks them and times them to the voice. You can still replace, move or hide any element.
Which format for publishing?
Vertical 9:16 MP4, the Explainer's output format. For training use, see also turning a course into explainer videos.
Written by the ClipOnIt team. Published November 7, 2026.