One idea in.
A finished short out.
AutoShortsX writes the script, generates every visual, records the voiceover, times the captions to the word, and renders a ready-to-post MP4 — from a single sentence. No editing timeline to learn. No render queue to wait in.
$197 $69 today Save $128
- Pay once, own it forever
- No watermark, no export limits
- 300+ voices included free
Four steps. You only do the first one.
Everything after your idea is automatic — and every step is still yours to edit.
Describe the video
Type an idea, choose a story type — fact, story, listicle, explainer — and pick an image style. That is the whole setup.
AI writes the script
Groq or OpenAI breaks your idea into scenes — one visual and one narration line each — paced for the length you asked for.
Visuals & voice fill in
Every scene gets an AI image, a stock photo or a stock clip, plus a voiceover in the voice you picked — generated in parallel.
Render and post
Ken Burns motion, transitions and word-timed captions are burned in by FFmpeg on your own PC. Out comes a finished MP4.
Six jobs. One app.
Scripting, art direction, voice acting, subtitling, motion and encoding — the entire chain a short normally takes, in a single window.
A script that already knows it is a short
Pick a story type and a length; the model returns a scene list, not an essay. Every scene is one visual and one spoken line, paced so the narration lands inside the shot instead of running past it.
- Story types: facts, storytelling, listicles, explainers, motivation
- Groq or OpenAI — your key, your model, at cost
- Rewrite any single line without regenerating the rest
Three ways to fill the frame
Generate original art, pull a stock photo, or drop in a stock video clip — chosen per project or swapped scene by scene when one shot is not working.
- AI images — Replicate, Runware.ai, Together AI, model-selectable
- Stock photos — Pexels and Pixabay, attribution tracked
- Stock video — real footage where a still will not do
300+ voices, none of them per-character
Voicely ships with the app: 300+ neural voices across 75+ languages, free forever, no key. Bring an OpenAI or ElevenLabs key when you want a specific performance — or your own cloned voice.
- Preview any voice before committing the whole video
- Rate and pitch control; steerable delivery on OpenAI
- ElevenLabs Multilingual v2, v3, Flash and Turbo supported
Captions that cannot misspell your script
Whisper listens to the generated audio for word-level timing — then every word is snapped back to the script you approved. Perfect timing, and the spelling is always the one you wrote.
- Word-by-word highlight, karaoke style
- Names and jargon never come out phonetic
- Font, size, colour and position all yours
He never missed a night
Stills that do not sit still
Every scene gets a Ken Burns move — a slow push, a drift, a pull back — with a transition into the next. Static slideshows stop scrolls; motion does not.
- Per-scene pan and zoom, direction and strength
- Transitions between every cut
- Scrub the whole timeline before you commit to a render
Rendered on your machine, not a queue
A bundled FFmpeg encodes every frame locally and muxes the per-scene audio, so picture and sound cannot drift. No upload, no credits, no waiting behind somebody else's export.
- 9:16 at 1080×1920 and 16:9 at 1920×1080
- Frame-accurate A/V sync by construction
- Render as many videos as you like, forever, for $0
Bundled free, or your own key.
Nothing is marked up. The voices are built in and cost you nothing — everything else runs on your own provider key, billed by them, at their price.
Voiceover
Voicely ships inside the app — 300+ neural voices across 75+ languages. No API key, no per-character cost, no cap. This is the one part you never pay anyone for.
AI script writing
Groq has a free tier, so you can write scripts at $0 from day one. Add OpenAI instead if you prefer their models — either way the key is yours and the bill is theirs.
AI image generation
Three providers, model-selectable. Costs land around 500 images per $1 on average, depending on the model you pick — paid straight to the provider, never marked up by us.
Stock photos & video
Both providers are free to use — you just register a free API key of your own. Real photography and stock footage for the scenes where generated art is not the right call.
Start completely free: bundled Voicely voices + Groq's free tier + free Pexels/Pixabay keys means your first videos can cost $0 beyond the licence.
Why not just use a cloud generator?
Because you end up renting your own output back, one credit at a time.
Cloud short-video sites
- Monthly fee, and it never stops
- Credits per video — run out mid-batch
- Watermark until you upgrade
- Their model, their voices, their limits
- Render queues at peak hours
- Your concepts sit on their servers
A desktop app you own
- One payment, then nothing, ever
- Unlimited videos — no credit meter
- Never a watermark
- Your keys, your models, provider pricing at cost
- Renders locally the second you hit go
- Keys encrypted in Windows Credential Manager
Built for people who post daily
Faceless channels
Run a whole niche channel without appearing on camera or recording a word. Batch a week of uploads in an evening.
Story & fact pages
History, true crime, space, mythology — pick the story type and let the script writer keep the hook in the first two seconds.
Agencies & freelancers
Turn one client brief into a month of shorts. No per-seat SaaS bill eating the margin on every retainer.
Multi-language creators
Same script, 75+ languages, native-sounding voices. Republish one video across every market you serve.
What early users say
I was paying $49 a month and still running out of credits by the third week. I made eleven shorts the first night I installed this and it cost me nothing after the licence.
The caption snapping is the feature nobody else got right. My niche is full of proper nouns and every other tool spelled them phonetically. This one just uses my script.
I produce shorts for six clients. Being able to render locally at 2am without a queue or a credit balance changed what I can promise on a deadline.
Questions, answered
What exactly does AutoShortsX do?
You type an idea and pick a story type and an image style. AutoShortsX writes a scene-by-scene script, generates or sources a visual for every scene, records a voiceover, transcribes it for word-perfect caption timing, animates the whole thing with Ken Burns motion and transitions, then renders a finished MP4 — all in one pass.
Do I need to know video editing?
No. The default run is fully automatic: idea in, finished short out. If you do want control, every scene is editable — swap the image, rewrite the line, change the voice, retime the captions, adjust the pan and zoom — but nothing is required.
Where do the visuals come from?
Three sources, selectable per project. AI-generated images via Replicate, Runware.ai or Together AI (you pick the model), stock photos from Pexels and Pixabay, or stock video clips from Pexels and Pixabay. Mix them scene by scene if you like.
What voices can I use?
Three engines. Voicely is bundled and free — 300+ neural voices across 75+ languages, no API key, no per-character cost. Add your own OpenAI key for gpt-4o-mini-tts with steerable delivery, or an ElevenLabs key for Multilingual v2, v3, Flash and Turbo plus your cloned voices.
Are the captions actually accurate?
Yes, and that is the point. Whisper transcribes each scene for word-level timing, then every word is snapped back to the script you already approved. If Whisper mishears a name or a technical term, the caption still shows the correct word with the right timing.
Is this a one-time payment?
Yes. Pay once and own it forever — no monthly subscription, no per-video render credits, no watermark, no export limits.
Does my work get uploaded anywhere?
Rendering happens entirely on your machine with a bundled FFmpeg — your video never touches a render queue. Script, image and voice generation call the provider you chose, and the free Voicely voices run locally too. Your API keys are stored encrypted in Windows Credential Manager, never on disk in plain text.
Do I need my own API keys?
Only for the AI parts, and they are cheap. Voicely voices are built in and free forever with no key at all. Script writing uses Groq (which has a free tier, so you can start at $0) or OpenAI. AI images use Replicate, Together AI or Runware.ai and work out around 500 images per $1 on average depending on the model. Pexels and Pixabay are free — you just register a free key. Every key is yours, so you pay providers directly at their price instead of a marked-up credit system.
Can it make horizontal videos too?
Yes. Both 9:16 vertical for Shorts, Reels and TikTok, and 16:9 horizontal for standard YouTube, work end-to-end through preview and render.
How long does one short take?
A typical 30-60 second short takes a few minutes end to end, most of it waiting on image generation. Rendering itself is local and fast — no cloud queue, no waiting behind other users.
What do I need to run it?
Any modern Windows 10 or 11 PC. FFmpeg and the voice engine are bundled — there is nothing else to install and no separate setup.
Your next short is one sentence away.
Pay once. Render forever. No subscription, no credits, no watermark.
$197 $69 Save $128
Windows 10 & 11 · FFmpeg and voices bundled · Nothing else to install