An AI explainer video in French needs three adjustments that English-first tools often skip: a shorter script (French takes 17 to 28% longer to say the same thing, measured on our five examples), a native French voice rather than an English voice "speaking French", and captions that follow French punctuation and number formats. The rest of the method stays the same: one question, the answer in the first seconds, one scene per idea. ClipOnIt's Explainer offers 6 French voices or your own voice on the mic, from the Starter plan ($29/month, 15 explainer videos a month).
Why French trips up AI video tools
Most explainer video generators are built in English and translated afterwards. It shows in three places:
- The voice: many "multilingual" voices are English voices trained to read other languages. The accent is fine for one sentence and noticeable over a minute, especially on liaisons and numbers.
- The pace: a script written for 45 seconds in English runs close to a minute once translated. Scenes stretch, the picture sits still, retention drops.
- The captions: compound words split badly, apostrophes cut at the end of a line, numbers shown in English format (1,000 instead of 1 000).
None of these stops you from publishing. Together, they make a video that sounds "translated", and a French-speaking audience hears it in the first sentence.
What French changes, in numbers
The five sample videos on the AI explainer video page exist in French and in English, with the same content and the same voice (the founder's). Here are their real lengths:
| Video | Words (FR / EN) | FR length | EN length | Difference |
|---|---|---|---|---|
| Where your coffee comes from | 145 / 135 | 56 s | 45 s | +25% |
| Why yawning is contagious | 98 / 94 | 37 s | 29 s | +28% |
| Why onions make you cry | 91 / 84 | 32 s | 26 s | +21% |
| How owls see at night | 65 / 69 | 26 s | 22 s | +17% |
| Why cats purr | 59 / 61 | 24 s | 19 s | +27% |
Measured on the published files on November 7, 2026. Even with the same word count, the French version runs longer: words have more syllables and the voice marks sentence endings more. On these five videos, French narration runs at about 150 to 160 words per minute, against 170 to 190 in English.
In practice: aim for 120 to 140 words for a 50-second video in French, and cut before you generate rather than after.
Example: "Where your coffee comes from", in French
This video was produced by the Explainer in the Satellite briefing style: the camera flies over the real Earth, from Ethiopia to the port of Le Havre. Voice-over: Mathis, ClipOnIt's founder (a private clone of his own voice, not offered to users). Below is the French version, the one this article is about.
The French narration, scene by scene (with an English gloss):
« Le café de ce matin a sans doute fait près de dix mille kilomètres avant d'arriver dans ta tasse. » (This morning's coffee probably traveled almost ten thousand kilometers before reaching your cup.) « Tout commence ici, sur les hauts plateaux d'Éthiopie : le berceau du caféier arabica. » […] « Éthiopie, Brésil, Vietnam : chaque tasse est un petit tour du monde. »
Three writing choices specific to French, visible in this script:
- Numbers written out ("dix mille", "soixante kilos"): the voice reads them without stumbling, and the picture shows the figure.
- Informal "tu", the norm on French TikTok and Reels; formal "vous" works better for training or corporate videos.
- Short sentences with colons: a long subordinate clause loses a French listener faster than an English one.
Writing a French script that sounds right out loud
- Write in French, don't translate. If you start from an English text, have it rewritten, not translated: English turns of phrase give the translation away.
- One idea per sentence, one sentence per scene. The Explainer splits the script into scenes; a 25-word sentence becomes a 10-second scene, too long for a single picture.
- Read it out loud before generating. Whatever trips your tongue will trip the synthetic voice.
- Avoid acronyms nobody says aloud, or spell them the way they're spoken.
- Keep the English loanwords your audience really uses ("short", "podcast") and translate the rest.
Picking the French voice
The Explainer gives you two options:
- One of the 6 French voices in the catalog: Guillaume (relaxed storyteller), Aslan (lively social-media tone), Cyp (soft, for explaining), Koraly (expressive storyteller), Victoire (young, TikTok tone) and Anaïs (clear, news tone). They are native French voices, not adapted English ones.
- Your own voice: record on the mic (one or several takes), import an audio file or even a YouTube link. The narration is transcribed word by word and each element appears when you name it. You can re-record a single sentence without redoing the video.
For a regular channel, the voice matters more than the visual style: it's what the audience recognizes. Pick one and keep it.
French captions: the details that matter
- 3 to 5 words per line, as in English, but French words are longer: use a condensed font or one size down.
- Never break a line after an apostrophe ("l' / oignon") or between a number and its unit.
- Accents on capitals ("À", "É") if captions are in uppercase: missing accents are noticed.
Caption styles (karaoke, boxed, outline…) are covered in our auto captions guide.
Check before you publish
- Watch the video once with the sound off: the captions alone should carry it.
- Listen to proper nouns and numbers: they're the first thing a synthetic voice gets wrong.
- Check every figure at the source. In the coffee example, "about a third of world production" for Brazil is a common order of magnitude; check yours.
- Cut the scene that drags. If a picture stays on screen for more than 8 seconds, the sentence is too long.
What it costs
The Explainer is included in ClipOnIt's paid plans: Starter at $29/month (15 explainer videos a month), Pro at $49/month (30) and Business at $149/month (100). The Free plan does not include explainer renders. Details on the pricing page.
FAQ
Can AI make an explainer video directly in French?
Yes. In the Explainer, you write the idea or the text in French, pick a French voice, and the narration, scenes and captions come out in French. Still reread the script: that step is where the quality comes from.
Should I write the script in English and translate it?
No, that is the most common cause of videos that sound off. Write directly in French; if you need both languages, rewrite the second version instead of translating it word for word, as we did for our examples.
How many words for a one-minute video in French?
Between 140 and 160 words, based on the pace measured on our examples (about 155 words per minute for the coffee video). For a 30-second short, aim for 70 to 80 words.
Can I make the same video in English afterwards?
Yes: duplicate the project and replace the text with its English version, using one of the 4 English voices in the catalog. Each language counts as one video in your quota. For the full method, see how to make an explainer video with AI.
Written by the ClipOnIt team. Published November 7, 2026.