Create a cartoon from text: you write the dialogue, the AI makes the clip
Describe the scene, name your characters, write their lines. Gemini Omni generates an 8-second clip in 3D or 2D, with the voices and the ambient sounds. One clip per scene; for a whole story, the AI Story tool chains the scenes for you.
Free account · a clip with dialogue from the Pack Mini, no subscription
From script to clip, without drawing or recording
Until now, a cartoon needed an illustrator, voice actors and an editor — or weeks in a piece of software. Here, the unit is the 8-second scene: you write it like a fragment of a script, the model draws the characters, animates them, has them say their lines and lays down the sounds. An episode is several scenes placed end to end — by hand with this tool, or in one go with AI Story, which keeps the same characters from scene to scene.
The tool is called text → video. With the Gemini Omni model, the sound and the dialogue are generated; with Kling Turbo, the video is silent but runs up to 10 seconds. If you are starting from a photo rather than from text, use animate a photo; for a single portrait that speaks with your voice, the talking avatar.
Three steps, five minutes
1. Write the scene like a script
The setting, the characters (looks, clothes, age), then the lines, one per row, with the name of who speaks: "KASSIM: Leo, come and look at this!". A clip lasts 8 seconds: two or three short lines, not a monologue.
2. Choose the style
Pixar 3D for the "animated film" look, comics or anime for 2D, watercolour or chibi for a softer tone. The style applies to the whole scene; keep the same one from clip to clip so the series holds together.
3. Generate, then chain the clips
Gemini Omni renders the scene in 8 seconds, with the voices of the lines and the ambient sounds, in under a minute. For a whole story — several scenes that follow on, the same characters from start to finish, voice-over and edit included — use the AI Story tool rather than chaining clips by hand.
A text → video clip made with ToonZap
No starting image: a written description, and the result exactly as the tool renders it.
What people make with it
The mini-series
The use that inspired this page: named characters, a neighbourhood, dialogue scene by scene — seven 8-second clips generated in a row, then cut into a one-minute episode. Since 18 October 2026, the AI Story tool does that work in one go, with the same characters from scene to scene.
The sketch or the dialogue
Two characters, an exchange of three lines, a punchline. The 8-second format is the format of the short sketch: it forces you to write tight, and that is what gets the laugh.
The explainer or the home-made advert
A character presenting a product, a mascot explaining a rule, an internal announcement: a cartoon gets across in 8 seconds what ten lines of text never will.
The children's story
A hero, a setting, one line per scene: a bedtime story in three or four clips, in the style the child likes best. The characters are invented from your description, not from a photo.
What it costs, no surprises
- An 8-second Gemini Omni clip, voices and sound included: 77 credits. A 5-second Kling Turbo clip, silent: 55 credits.
- The free account comes with 40 credits (three photo → cartoon conversions). The Pack Start (€1.99) pays for a 5-second Kling video; the Pack Mini (€4.99, 150 credits) for one Omni clip and one cartoon.
- The Pack Plus (€14.99, 550 credits): seven 8-second clips with dialogue — almost a minute of episode. No subscription, and the credits never expire.
- A refusal (protected character, sensitive content) or a failure is refunded automatically.
A whole story rather than a clip? AI Story
This page is for the single clip: one scene, a sketch, a home-made advert. To tell a story of 30 seconds to 3 minutes, the AI Story tool writes the script from your idea, fixes one visual sheet per character and reuses it in every scene, chains the scenes (same setting, same light), records the voice-over, adds subtitles and music, and delivers the edited video — in an illustrated, animated or cinema look (where the characters speak their own lines, lips in sync). Nothing left to paste from clip to clip, no edit left to do.
Free account · script reviewed before you pay in Step-by-step mode
Everything the tool does, its prices and its limits: create an AI story video.
Frequently asked questions
How much does a clip cost, and a whole episode?
An 8-second Gemini Omni clip, with sound and dialogue generated, costs 77 credits. The free account comes with 40 credits (three photo → cartoon conversions), not enough for a clip. The Pack Start (€1.99, 55 credits) pays for a silent 5-second video with Kling Turbo; for an Omni clip you need the Pack Mini (€4.99, 150 credits: one clip and one cartoon); the Pack Plus (€14.99, 550 credits) covers seven 8-second clips — almost a minute of episode. No subscription is needed, and the credits never expire.
Are the voices really generated from my dialogue?
Yes: Gemini Omni generates the scene, the voices of the written lines and the ambient sounds in a single 8-second pass. They are synthetic voices chosen by the model — you do not pick the timbre, and a dialogue too long for 8 seconds will be cut or sped up. Write two or three short lines per clip.
Will my characters stay the same from clip to clip?
With the text → video tool, each clip is generated separately: from one clip to the next, a character can change face or clothes, even when you paste the same description. For a story where the characters stay the same from scene to scene, use the AI Story tool: it fixes one visual sheet per character and reuses it in every scene, chains the scenes (same setting, same light) and delivers the edited video with voice-over, subtitles and music, from 30 seconds to 3 minutes.
Can I use well-known characters (Simpsons, Disney, Marvel)?
No. The models refuse protected characters, and ToonZap warns you before sending when it spots one in your text. Describe your own characters — a style "in the manner of" (anime, 3D, comics) is possible, a licensed character is not. A refusal is refunded automatically.
What length, format and quality?
8 seconds per clip with Gemini Omni, in 720p, MP4 export. With Kling Turbo, 5 or 10 seconds without sound. Choose the format before you start: 16:9 for YouTube or an edit, 9:16 for TikTok and stories.
Your first scene is waiting
Create your account, write two lines and a setting, pick a style: the clip is ready in under a minute.
No credit card to sign up · 8 seconds with voices for 77 credits