GUIDES · 2026
Turning a short written story into a video usually means one of two things: paying for an AI video generator, or spending hours in a full animation program. Our free 2D Story Video Maker takes a third path — you write your story scene-by-scene, pick from a built-in library of flat-style illustrated backgrounds and characters, add narration, and export a real downloadable video. No paid AI API, no subscription, no signup — everything runs in your browser.
This is not an AI image generator. It doesn't paint a brand-new picture for every scene the way a diffusion model would — that technology needs a paid API to run, and this site's whole approach is to stay free forever. Instead, the tool ships with a hand-built library of consistent flat-style illustrations (think simple, clean, modern icon art) — 10 backgrounds and 8 "peg doll" style characters — and your job is to assemble a story from that library, the way you'd lay out a picture book. Once you see it that way, the workflow makes a lot more sense.
Open the Story Video Maker and you'll see 3 scenes already started. Each scene is one beat of your story — one setting, one moment. Type the narration text for a scene directly into its text box: this is both what gets spoken (or captioned) and what the auto-suggest feature scans to guess a matching background and characters. Add more scenes with "+ Add Scene" (up to 15), and reorder or delete them with the ▲ ▼ and 🗑 buttons in each scene's header.
Click "✨ Auto-Suggest Background & Characters" on a scene and the tool scans your narration text for keywords — words like "school," "village," "forest," "beach," "night," "dog," or "grandmother" — and pre-selects a background and up to two characters that seem to match. It's a helpful starting point, not a mind-reader, so always glance at the pick and override it manually if you want something different: click any background thumbnail in the grid to select it, and use each character slot's grid plus its left/center/right position dropdown to place people (or a dog, or a cat) exactly where you want them in the frame.
This is the part worth reading carefully, because it's the one real technical limit of browser-based video tools. No mainstream browser lets a website capture the audio a synthesized (text-to-speech) voice produces into a recordable track — there's simply no reliable API for it. So the tool gives you two honest, clearly separated paths instead of pretending to combine them:
Speak each scene's narration into your mic. This is the only path that produces a downloadable video with real audio, because recorded audio (unlike synthesized speech) can be routed through the Web Audio API into a capturable stream.
Type narration text and preview it live with a computer voice. The exported video from this mode is silent, with your narration burned in as on-screen captions instead — genuinely useful, just not the same as audio.
If you want a narrated, downloadable video, use "Record My Own Voice." If you're happy with a captioned, silent video — or you just want to draft and preview your story quickly — "Type & Preview with TTS" is faster. You can also combine the two: use TTS preview mode to test pacing, then switch to Record mode and read the same lines into your mic once you're happy with the story.
Switch to the "🎙️ Record My Own Voice" tab and each scene grows a "🎙️ Record Narration" button. Click it, allow microphone access the first time your browser asks, speak your line, then click "⏹ Stop." An audio player appears right there so you can check the take, and re-record any time before you export — a scene's duration automatically follows its recorded clip's real length unless you type a manual override into the duration field.
Every scene has a "Show captions on screen" toggle, which burns the narration text onto the video as on-screen captions — turn it on for accessibility, for silent-scroll social feeds, or as the whole point if you're using TTS-preview mode's silent export. Each scene's duration field is editable too: in Record mode it defaults to your recorded clip's length, in TTS mode it's estimated from your text at roughly 150 words per minute (minimum 3 seconds) — type your own number any time to override either estimate.
Click "▶ Preview Story" to watch the whole animated sequence play through in the canvas — backgrounds gently zoom and pan (a soft Ken-Burns effect), characters ease in with a slide-and-fade at the start of each scene and bob gently while they're on screen, and in TTS mode you'll hear each scene's narration spoken aloud in sync. Use "⏹ Stop Preview" to interrupt it early once you've seen enough.
Click "⬇ Download Video" when you're ready. This isn't instant — the tool plays through your entire story in real time while it records the canvas (and, in Record mode, mixes in your recorded audio), because that's fundamentally how browser-based capture works. A progress message tells you exactly which scene is being rendered ("Rendering scene 2 of 5…"), so keep the tab open and active until it finishes. When it's done, you'll get a standard .webm video file — the same format this site's Screen Recorder and Voice Recorder use — ready to save, share, or convert to MP4 with any free converter if you need that format.
Every illustration is pre-built; there's no per-image generation cost and nothing metered.
Recorded voice, story text, and the exported video are all created and kept on your own device.
Not a slideshow — a genuine .webm export you can upload, share, or convert.
Ready to try it? Head over to the 2D Story Video Maker and turn your next story idea into a shareable video in a few minutes.