Guide
How to make a stick figure video from a script with AI
Paste the script into a tool that both narrates and draws it, such as Alpy, and pick a stick figure character. An AI voice reads the script, Alpy draws a simple stick figure scene for each part of it, and every drawing cuts in on the words it shows. You get a finished 16:9 YouTube video with the same stickman in every scene. Scripts need 110 to 2,700 words, and a free preview shows the first 10 seconds before you pay.
Last updated 15 October 2026
Try Alpy with a free preview →How to do it, step by step
- Write the script for listening. Short sentences, one idea each, with the interesting part first. Every sentence becomes something to draw, so concrete beats abstract: "he checks his phone 96 times a day" draws better than "modern habits".
- Paste it into the studio. In Alpy's studio, choose AI voice and paste the script. A reply from ChatGPT or Claude can go in whole: the title, labels such as "Hook:", notes in [brackets] and stage directions such as "(pause)" are left out of the narration, and the studio shows what will not be read. Scripts need at least 110 words, a little over 30 seconds.
- Pick a voice and a delivery. Six AI voices, each in 4 deliveries, from calm to high energy. Each voice has a sample to listen to before you choose.
- Pick a stick figure. The Stickman, the Classic Stickman, or the Ancient Stickman for history and prehistory. All are free, and you can upload your own drawing instead.
- Watch the free preview, then finish. The first 10 seconds are narrated, drawn and cut before you pay. Finish the whole video, redraw any scene you do not like (the first five redraws on a video are free), and download the MP4 with three title ideas, a description and a thumbnail.
The voices and deliveries
| Voice | Sounds like |
|---|---|
| Ash | Warm, steady male |
| Coral | Bright, friendly female |
| Onyx | Deep, confident male |
| Sage | Calm, thoughtful female |
| Nova | Clear, upbeat female |
| Ballad | Warm British storyteller, male |
Each voice reads in one of these deliveries: energetic explainer, calm and clear, storyteller, high energy. A calm read comes to about 140–150 words a minute and a high-energy one to about 180–210, so the same script makes a shorter video at high energy.
What a finished one looks like
- Why You Can't Tickle Yourself: a 2:24 science explainer with the Stickman, read by Nova as an energetic explainer, 28 drawn scenes.
- Ancient Humans Never Brushed: a 2:52 history explainer with the Ancient Stickman, read by Onyx as a storyteller, 33 drawn scenes, one redrawn in the studio.
Both were made from typed scripts in the studio, the way any customer's video is made. They are Alpy's own examples, not customer videos.
Ways to make a stick figure video
| Way | What you do | What you get |
|---|---|---|
| Alpy | Paste the script, pick a voice and a stick figure | Narration and a drawn scene for each part of it, in minutes |
| Hire a stick figure animator (Fiverr, Upwork) | Send the script and a voiceover, wait for delivery | Custom animation; price and turnaround vary by seller |
| Animate it yourself (Pivot Animator, Adobe Animate) | Pose and time every frame by hand, record or generate the voice separately | Full control, many hours per minute of video |
| Script-to-video AI tools (invideo AI, Fliki) | Paste the script and pick a voice | Stock footage or generated clips, rarely a consistent stick figure |
The rows describe how each way works with a written script; see each service's own site for current prices.
Mistakes to avoid
- Pasting a script nobody has read aloud. An AI voice reads exactly what is there. Read it once yourself and cut what trips you up.
- Long paragraphs of abstraction. Scenes change every few seconds, at the ends of sentences where one is close; a paragraph with nothing to show gets vague drawings.
- Numbers you have not checked. The drawings show what the script says, so a wrong figure ends up on a sign in the video. Check facts before you generate, and redraw any scene that gets one wrong.
- Expecting Shorts. Alpy makes horizontal 16:9 videos for YouTube, not vertical Shorts or Reels.
Where Alpy is not the right fit
- Six AI voices, not a large voice library, and no voice cloning
- Horizontal 16:9 only, no vertical Shorts
- One deliberately simple drawing style, not polished animation
- No free plan beyond the preview
Questions
- Is there an AI that makes stick figure videos from a script?
- Yes. Alpy takes a typed script, reads it with one of six AI voices, and draws a stick figure scene for each part, with the same character throughout. The first 10 seconds are a free preview.
- Can I use a script from ChatGPT?
- Yes. Paste the whole reply: titles, section labels, bracketed notes and stage directions are left out of the narration, and the studio shows what will not be read before you start.
- Can I use my own voice instead of an AI voice?
- Yes. Record or upload yourself reading the script and Alpy draws the video around your recording. Alpy never clones a voice.
- How long can the video be?
- Scripts can be 110 to 2,700 words, and videos up to 20 minutes. About 150 words make a minute of narration at a calm pace.
- Can stick figure videos be monetised on YouTube?
- Faceless channels can join the YouTube Partner Program. What YouTube excludes is mass-produced or repetitive content, so original scripts with something to say are what matter, whoever reads them.
- How much does it cost?
- The preview is free with no card. Finishing one video is $7 a video or $29/month.
Related
Sources
Other tools’ prices and features were checked on their own sites on the date above and change often. Spotted something out of date? Tell us at hello@tryalpy.com.
See your own video before you pay.
Record a few sentences in your voice and Alpy illustrates the opening for free. No card.
Make my free preview →