Open source · self-hosted or hosted

Turn a topic into a ready-to-post vertical video.

Script, voiceover, word-synced captions, visuals and music — from a single topic. Or bring your own: caption or dub a video, cut a long one into clips, turn your clips into an edit cut to a song, and post to YouTube automatically. Pay per video or self-host it for nothing; no subscription either way.

languages
10
languages
art styles
7
art styles
free credits
50
free credits
licensed
MIT
licensed
FollowingFor You

@shortpulse

Why honey never spoils 🍯

original sound — shortpulse

SH
stock_media · en · 24s · real render
New format · What if?

Ask a question, get a video people can't scroll past

Type a question like "What if the Moon vanished?" and it comes back scripted from checked facts, with glowing maps, a counter and a growing list, hard cuts and a narrator.

  • Glowing maps drawn from real borders — the country goes out when it vanishes.
  • Facts are gathered and checked first; the script is written only from them.
  • A population counter, a list that grows with every consequence, chapter titles and stamps.
  • A narrator in your language, dramatic pauses and sound effects.
Examples:
New

Two things you haven't seen it do yet

Both of these came out of ShortPulse as they are. Press the buttons.

Clips in, an edit cut to the beat out

Drop in your clips and a song. It finds the beat and the drop, shows every move from start to finish, rushes into the drop and opens on your title.

9 clips + 1 song
Winning Spirit · 123 BPM
Style

One video, captions in any language

Upload once, pick the language the captions should be in. The words are translated and timed to the voice — the audio stays yours.

1 video, spoken in English
An octopus has three hearts. Two of them pump blood through its gills. The third keeps the rest of its body alive. But when an octopus swims, that third heart stops beating. Which is why it would rather crawl.
Captions · English
Captions

Footage from Pexels, filmed by ArtHouse Studio, Tom Fisk, Alex Moliski, cottonbro studio, Luke Cromwell, Marc Espejo, Collab Media, Italo Bicca, Entdecker Fuchs, JUN HO LEE. Music: “Winning Spirit” (Pixabay), “Calm Currents” by HoliznaCC0. Narration: Piper.

Real output

Nineteen videos this pipeline actually made

Unedited output — AI-written script, Piper voiceover, word-synced captions, in ten different languages. Eighteen use real stock footage; the one marked AI stills was drawn by the paid mode. Hover any one to play it.

English
Why honey never spoils24s · 5 scenes
Türkçe
Neden gece rüya görürüz?20s · 5 scenes
Polski
Dlaczego miód nigdy się nie psuje?24s · 5 scenes
Français
Pourquoi la lune change-t-elle de forme ?25s · 6 scenes
English
Why the ocean is salty19s · 5 scenes
Deutsch
Warum färben sich Blätter im Herbst?20s · 5 scenes
Español
¿Por qué el cielo es azul?17s · 5 scenes
العربية
لماذا تخاف القطط من الماء؟21s · 5 scenes
English
How lightning actually forms23s · 5 scenes
Português
Por que bocejamos?26s · 5 scenes
English
Why coffee wakes you up24s · 5 scenes
Русский
Почему небо ночью тёмное?11s · 5 scenes
Türkçe
Kediler neden kutuları sever?20s · 5 scenes
Polski
Dlaczego ocean jest słony?21s · 5 scenes
Italiano
Perché il caffè ci sveglia?16s · 5 scenes
English
Why volcanoes erupt24s · 5 scenes
English
Why we get goosebumps18s · 5 scenes
English
A simple trick for better sleep19s · 5 scenes
EnglishAI stills
Why quiet people read the room24s · 8 scenes

Footage from Pexels, filmed by Aaron Burden, Abdullah | 4K, Adventure Studio, Aleks Magnusson, Alessio Casaretto, Alex Moliski, Alexey Chudin, Alvin David, Ambareesh Sridhar Photography, Ana Sandu, Andre Moura, Andres Perez, Angela Roma, Anna Pou, Anna Shvets, ANTONI SHKRABA production, Artem Podrez, ArtHouse Studio, Ashad The Nomadic, aslı aydoğdu, Bahri Gün, Barbara Olsen, Bav Vadgama, Ben Prater, Canan İldeniz, cottonbro studio, Darina Belonogova, Deti riyanti, Ebahir, Emrah, Engin Akyurt, Eyüp Can, Grigoriy Bunkov, Hale Ş, Hashim Suhimi, Iceberg San, Ilya Lyzhin, İrem Meriç, John Diez, Joolsmagools ®️, Juan Camilo Trujillo Botero 🇨🇴📸, JUN HO LEE, K, Kakada Chuon, Kevin Malik, Kevin Villanueva, khezez | خزاز, Koushalya Karthikeyan, LauraB, Lentes Bella, Los Muertos Crew, Luis Quintero, Marc Espejo, Marina Leonova, Masha Glazova, Matthias Groeneveld, Max Medyk, Md Amir Umar, Micaela Alejandro, Michael Burrows, Mikhail Nilov, Miriam Alonso, Mizuno K, muhammed diler, Muhtelifane, Nadezhda Moryak, Nataliya Vaitkevich, Nereid Ndreu, Nicola Narracci, Nikita Ryumshin, Nisasu, NQPPresents, Nur, Pachon in Motion, Pavel Danilyuk, Photoviewx, Physical Pixel, RDNE Stock project, ROMAN ODINTSOV, Ron Lach, Sandeep Solanki, Sema, Shan Ali, Stefanie Jockschat, Thuan Pham, Tima Miroshnichenko, Timothy Fuller, Timur Weber, Tom Fisk, Toni.063371 - Antonio Sáez, Yuliya Duzhaya, Şahin Doğdu. The AI-stills clip has no footage to credit — every frame of it was generated.

Background music is “Airport Lounge” by Kevin MacLeod (incompetech.com), licensed CC BY 3.0 — the same royalty-free track the pipeline falls back to by default.

The whole studio

Six jobs, one place to do them

Start from a topic or from footage you already have. These are screens of the app itself — pick a tab to see each part.

The studio with a topic typed in, the settings rows and a live caption previewThe caption and dub tab, set to dub into Turkish, with the caption templates openThe library with a source video, the three clips cut from it, an edit and a captioned uploadThe edit tab with clips chosen, a song's waveform with its drop marked, the style and an on-video titleThe editor with the caption lines of a finished video beside its preview

Type a topic. It writes the script, records the voiceover, finds footage or draws stills, times every caption and mixes in music — one finished vertical video.

New

One long recording, several posts

Drop in a podcast or a talk, choose the stretch worth reading, and the transcript picks the moments that stand up on their own. Each one comes out captioned, in your library, ready to edit.

Up close: captions light up word by word, in step with the voice.
Cut it into clips
Don't — keep it whole2 clips3 clips4 clips5 clips
Which part to use
4:00→7:308:00

8 credits · 3:30 read

What came out
Documentation Quality0:15 · 9:16Captioned
Speed and Clarity0:12 · 9:16Captioned
  • The transcript picks the moments — three were asked for here and two were there.
  • Only the stretch you choose is transcribed, so an hour of recording is not an hour of work.
  • Priced by what was read, plus each clip: one credit a minute, and the price of a caption job per clip. Clips that aren't found are given back.
  • Every clip lands as its own project — captions, frame and edits included.

Built on open source you can audit

Ollamalocal LLM runtime
Llama 3scene scripting
Piperneural voiceover
faster-whisperword-level timing
Stable Diffusion XLimage generation
RealVisXLphotoreal stills
LTX-Videotext-to-video
FFmpegassembly & encoding
libassburned-in captions
Pexelsstock footage
FastAPIrender API
Next.jsthis interface
How it works

Three decisions, then it runs itself

1

Give it a topic

One line is enough — the LLM writes the scene breakdown, the hook and the call to action. Already have a script? Paste it and the AI only splits it into scenes.

2

Pick how it looks

Language, length, visual engine and caption style. The exact credit cost and an estimated render time are shown before you commit to anything.

3

Watch it build, then post it

Progress streams live over a WebSocket, stage by stage. You get a 1080x1920 MP4 with burned-in captions and ducked music — yours to download, no watermark.

Art styles

Seven looks, one prompt away

Every frame below is a real render — same model, same settings, same subject. We picked a subject with a face and hands on purpose, because that's the hardest thing to get right and the reason most of these looks are stylized.

PhotorealLooks like footage. Best for objects and scenery, worst for people.
AnimeCel shading, clean line art, saturated colour.
3D ToonBig-eyed animated-film characters, soft light.
ComicInked outlines and halftone shading.
ClaymationPlasticine models with visible fingerprints.
Flat CartoonFlat-colour explainer drawings with clean outlines.
Vector StoryThick outlines, flat muted colours, simple faces: the serialised-story look.
Pixel ArtChunky 16-bit sprites, limited palette.
Caption styles

The part that actually holds attention

Captions are timed per word by faster-whisper, so the active word highlights on the syllable it's spoken. Pick how that looks — size, colour, placement and how many words sit on screen at once.

your words land on the beat

The editor

Every scene, lined up against the video

When a render finishes you get the whole breakdown next to the player — what the voice says, how long each scene runs, and where it sits in the timeline.

Scene breakdown

Playing

Hook: Honey, the ultimate superfood, is surprisingly eternal!

Scene 14.0s

It's because of its unique acidity level.

Scene 24.0s

Honey's pH level is naturally acidic, around 3.2 to 4.5.

Scene 33.0s

This acidity creates an environment where mold and bacteria can't thrive.

Scene 44.0s

Additionally, honey's sugar content is so high that it dehydrates any microorganisms.

  • Click any scene to jump the video to it
  • The playing scene highlights itself as it goes
  • Per-scene timing, straight from the word-level transcript
  • Footage credits collected for you, with a copy button
Pipeline

Five local stages, one finished .mp4

01

Script

Ollama (local) or OpenAI breaks your topic into scenes

02

Voiceover

Piper voices, fully local, 10 languages

03

Captions

faster-whisper times every word for karaoke-style burn-in

04

Visuals

AI stills or free stock footage; local text-to-video when you self-host

05

Assemble

FFmpeg mixes ducked music and renders the final .mp4

What's built in

Everything short-form video needs, none of it gated

Clips in, an edit out

Drop in your clips and a song; it cuts them on the beat in one of three styles, keeps every move whole, zooms into the drop and can open on a title of yours.

Long video in, clips out

A podcast, a talk or a match reel goes in; the moments worth posting come back as clips — chosen from what is said, or from the picture when little is. Priced by the minutes read plus each clip.

Dub into another language

Your own video, spoken again in any of the ten languages over the original picture — with captions to match.

Posts to YouTube for you

Connect a channel once and every finished render can go out automatically. Instagram and Facebook are coming.

What's rising, as ideas

Pick a region and see what is climbing on YouTube and in the charts this week, turned into topics and edits this studio can actually make.

Bring it in from a link

Paste a Google Drive, Dropbox, OneDrive or Box share link instead of uploading — the file lands in My files, ready to caption, split or edit.

Runs on your machine

Self-host it and scripting, voiceover, transcription, visuals and rendering all execute locally — only stock footage touches the network. The hosted service trades that away deliberately: it has no GPU, so the script and the AI stills come from paid APIs.

Word-synced captions

faster-whisper gives word-level timing, so captions highlight one word at a time instead of dumping a full line.

10 languages

Script, voiceover and subtitles generate natively in English, Turkish, Spanish, German, Arabic and more.

Auto-ducked music

Background music sidechain-ducks under the voiceover automatically — no manual mixing.

Photoreal stills or real footage

A RealVisXL still per scene with a Ken Burns move over it, or a matching Pexels clip. Self-hosting with a GPU adds local text-to-video, which this service doesn't run.

Branded outro card

An optional closing card rendered locally with Pillow, so the last frame is your call-to-action, not a diffusion guess.

Why not just subscribe

What changes when it's local and open

What you payShortPulse: Per video, or nothing self-hostedTypical subscription tool: $20–50/mo, posted or not
Idle monthsShortPulse: Cost nothingTypical subscription tool: Billed anyway
Source codeShortPulse: MIT, fully readableTypical subscription tool: Closed
Where it runsShortPulse: Your machine, or oursTypical subscription tool: Their servers only
CustomizationShortPulse: Edit prompts, swap models freelyTypical subscription tool: Locked to their pipeline
Unused creditsShortPulse: Never expireTypical subscription tool: Reset every month
Pricing

Pay for videos, not for a month you didn't use

Credits are one-off and never expire. No plan renews, nothing charges you again unless you buy again — and the whole thing is still MIT if you'd rather run it yourself.

Free

Every new account, no card.

$0
  • 50 credits on sign-up
  • Your first AI-stills video free
  • Then stock-footage videos from 2 credits
  • Every language
  • No watermark

Starter

Testing the water on a posting habit.

$11.90one-off
  • 100 credits
  • ~16 standard videos
  • ~50 with stock footage
  • Never expire
Most picked

Creator

A daily short with room to redo the ones that miss.

$34.90one-off
  • 400 credits
  • ~66 standard videos
  • ~200 with stock footage
  • Never expire

Studio

Multiple accounts, or a client workload.

$94.90one-off
  • 1200 credits
  • ~200 standard videos
  • ~600 with stock footage
  • Never expire

Or pay nothing at all

Self-host it and the price is zero, forever — credits only exist because the hosted instance runs renders on hardware someone has to pay for.

2 credits = stock footage · 6 = AI stills · 40 = animated (every scene moves), each ×2 for medium and ×3 for long. You're quoted the exact cost before a render starts.

Prices in USD. Cards from any country work — your bank converts at its own rate.

Built to survive scrutiny

Marketing pages hide the rough edges. Ours are in the README, in full — here are a few of them, so you know before you clone it:

  • AI stills draw faces convincingly and extremities badly — hands, feet and full-body shots are where a frame falls apart, so the script engine keeps people in close and medium shots. Legible text is beyond the model entirely: it cannot write a label, a sign or a book cover.
  • Animated scenes are AI stills brought to life by an image-to-video model at 480p and upscaled, so they are softer than the stills. When you self-host, this mode runs local text-to-video instead and needs a real GPU.
  • Small local LLMs occasionally under-count scenes; the script engine retries and drops malformed ones rather than failing the render.
  • Stock footage is a closest-match, not a guarantee — Pexels clips can be loosely related to the scene.
FAQ

The questions worth asking first

Videos

What exactly do I get?

A 1080x1920 MP4, H.264/AAC, with the voiceover, word-synced captions burned in, background music ducked under the voice, and an optional closing card. No watermark, on every plan including the free credits.

How long does a render take?

On this service, a short with stock footage takes about 45 seconds end to end. AI stills take one to two minutes: the images come from a hosted GPU at roughly 8 seconds each, but the first one after a quiet spell waits for the model to load. Self-hosted on an Apple Silicon M-series the stills are much slower — about 78 seconds per scene, so ~7 minutes for a short — because your own machine is doing the diffusion.

Which languages are supported?

Ten: English, Turkish, Polish, Spanish, French, German, Portuguese, Arabic, Russian and Italian. Script, voiceover and subtitles are all produced in the language you pick. Japanese is missing because Piper ships no Japanese voice we can use commercially.

Can I use my own script?

Yes. Paste it and the LLM only splits it into scenes and writes the visual prompts — your wording reaches the voiceover untouched.

Do you post to TikTok or YouTube for me?

No. ShortPulse renders the file and hands it to you; publishing is still your job. If auto-posting is the feature you actually want, another tool will serve you better today.

Do I own what I make?

Yes. Download it and do what you like with it. When you use the stock-footage mode, credit the videographers in your post description — the project page collects the names and links for you, and Pexels' licence requires it.

Credits & billing

How does pricing work?

Credits, bought once. A render costs 2 credits with stock footage and 6 with AI stills, doubled for medium length and tripled for long. A beat edit is 3 credits up to 15 seconds, 5 up to 30 and 8 up to 60. You see the exact cost before you start it. (Local text-to-video is 20, and only applies if you self-host — it is not offered here.)

Do credits expire?

No. Nothing renews and nothing resets monthly — the ledger has no expiry, so credits sit there until you spend them.

What if a render fails?

The credits go back automatically. Refunds are tied to the project and can only happen once, so a failure can't leave you charged for a video you never got.

Is there really a free tier?

50 credits when you sign up, no card. One of your videos can use AI stills — that one is on us, so you see the paid mode before deciding anything — and the rest of the grant renders stock-footage videos at 2 credits each, or beat edits and captioned uploads. Re-rolling a scene and further AI-stills renders need a credit pack.

Self-hosting

Can I run this myself instead?

Yes, and it costs nothing. The whole thing is MIT — clone it, point it at your own Ollama and GPU, and there are no accounts, no credits and no limits. Credits exist only because the hosted instance runs on hardware somebody pays for.

Is it genuinely local?

When you self-host, yes: scripting, voiceover, transcription, image generation and rendering all run on your machine, and the one stage that reaches the network is stock footage from Pexels. This hosted service is the trade-off — it has no GPU, so scripts and voiceover come from OpenAI, AI stills and animated scenes from Replicate, and the speech in uploads longer than three minutes is transcribed by OpenAI. Shorter captions and rendering still happen on our server, and the code is the same either way.

What are the rough edges?

AI stills draw faces well and extremities badly — hands, feet and full-body shots are where a frame falls apart — and legible text is beyond the model entirely. Local text-to-video needs a serious GPU, which is why this service doesn't offer it at all. Small local LLMs sometimes return fewer scenes than asked. All of it is in the README's Known limitations section.

Start with 50 credits. No card, no trial timer.

That's enough to judge it by. Buy more only if it earns it — or clone the repo and run the whole thing on your own hardware for nothing.