You take the helm.
Your AI crews the busywork.

VidHelm is a free, open-source desktop video editor built to be flown with an AI co-captain. Cut the dead air, narrate in one take, drop sound effects on your beats, spin your 3D prints over your footage, and export YouTube-ready, by hand, or by asking.

⬇ Download for WindowsView source on GitHub

Free & open source (MIT) · Windows beta · everything runs on your machine, no accounts, no uploads

The VidHelm editor: media bin, preview with a text overlay, timeline with filmstrip and tag points, and the export panel
Read this first: VidHelm is not a standalone editor yet. On its own it cuts, tags, narrates and exports just fine. The parts that make it worth using, writing your titles, placing effects on every beat, running your whole workflow from one sentence, need an assistant connected. Setup is one command and takes about a minute, and the app has a panel that health-checks it for you.
0 accounts0 uploads13 synthesized SFX25 AI tools−14 LUFS mastered exports1 take narrationSTL / 3MF / OBJ on the timeline

An editor built around how makers actually work

Import → tag the beats → tighten → narrate → sound → export. Every step has a fast path.

✂️

Cut Pauses, seamlessly

Finds every silent gap, or motionless stretch in silent footage, and splices it out with tiny crossfades. A rambling take becomes a tight one in one click.

🏷

Tag points

Tap M at every beat, the joke, the reveal, the pop. Tags are the language of the whole app: clips snap to them, SFX land on them, narration lines sync to them.

🎙

Karaoke booth

Paste your script and hit record: the video plays while your lines light up in time. One continuous take, no clicking between lines, lands straight on the voice track.

🗣

Cloned-voice narration

A setup wizard records about twenty seconds of your voice and installs a free local engine for you: XTTS-v2, or audio.cpp if you want no Python and licence-clean models. Lines land on your tag points automatically.

🔊

Sound effects, three ways

Thirteen classic effects synthesized on your machine, your own sounds dropped into a folder, or press record and make one yourself. Describe a sound in words and a local model can generate it too.

🧊

3D Studio

Drop in an STL, 3MF, OBJ or GLB, pose it, pick a colour and finish, and render a spinning turntable straight onto the timeline. Pick the green screen backdrop and your print floats over your footage.

📁

Project folders, no importing

Point VidHelm at one folder and every sub-folder inside it becomes a project. Dropping files in with Explorer is the import. Saving writes back to the same folder, so a project is something you can copy or hand to someone.

📝

Auto captions

On-device Whisper transcribes the whole timeline into styled captions, phrase or word-by-word karaoke mode. Nothing leaves your computer.

🎚

Audio done right

Labeled voice / SFX lanes, drawable volume automation, and one checkbox that masters the mix to YouTube's −14 LUFS target with peaks in check.

✅

Watch & Verify

Every export is auto-checked, loudness, true peak, black frames, sampled stills, so you never upload a broken render again.

🖼

Thumbnail builder

Sample frames from your video, pick one, type a catchy subtitle, out comes a 1280×720 thumbnail with your logo composited on.

📥

Imports just about anything

Video, audio and image formats are decided by FFmpeg rather than a fixed list, so unusual files still work. Anything unusable is refused with a plain-English reason instead of landing as a broken clip.

VidHelm 3D Studio panel with a multicolour 3D printer model loaded, colour and finish pickers, backdrop options and turntable render controls
3D Studio: pose a model, keep its own colours, render a turntable, key out the backdrop
VidHelm Connect your AI panel showing green MCP health checks and per-client setup instructions
The AI panel health-checks the connection and writes the config for your client

The part nobody else has

VidHelm ships an MCP server. Point Claude (or any agent) at it and your AI gets 25 tools to drive the running app, while you watch it happen in your window.

I tagged all the beats. Make it fun and tighten it up.
On it, cutting 18 pauses (14s of dead air)…▸ cut_pauses · place_sfx("party", t=reveal) · add_text("wait for it…")
Done, here's a screenshot. Want the horn louder?▸ screenshot · export_video → ✓ YouTube-ready (−14 LUFS)
  • Same timeline, live. The AI's edits appear instantly in your GUI, and Ctrl+Z works on them.
  • Tag points are the shared language. You mark the beats; it hangs SFX, titles and narration on them.
  • It can see. A screenshot tool shows your AI exactly what you see, and every export returns a quality report.
  • Bring your own AI. Zero config for Claude Code, Cursor and VS Code. Copy-paste setup for Claude Desktop, Windsurf, Cline, Codex CLI and Gemini CLI. Fully local with LM Studio, Jan, Open WebUI, Ollama or AMD Lemonade Server, and a plain HTTP bridge for everything else.
  • Stuck? The app fixes it. The AI panel runs live diagnostics, generates the config with your real install path, and walks you through the usual snags.

Start Recipes, like start G-code, for videos

Your standing instructions, run on every new video. Toggle steps on and off (off = #-commented, just like G-code), free-type anything, your AI reads it and does the busywork.

cut-pauses# tighten dead air before anything else
thumbnail# sample frames, pick one in the picker
subtitle# catchy one-liner on the thumbnail
titles 5# AI pitches 5 title options, you pick
logo bottom-right# brand watermark on every export
intro-audio# your intro sting at 0:00
# captions# toggled off, just flip the chip

An AI video editor that cuts out pauses for you

The slowest part of editing a talking-head video is hunting down every "um", breath and dead-air gap. VidHelm does that pass in one click, or you can just ask your AI to do it.

Automatic silence removal

VidHelm scans the whole timeline for silent gaps and, in silent footage, motionless stretches, then splices them out with short crossfades so the cuts stay seamless. A twenty-minute rambling take turns into a tight edit before you touch anything else.

Free AI video editing software

No subscription, no watermark, no render credits. VidHelm is MIT-licensed and open source, ships as a Windows installer with FFmpeg bundled, and every AI feature, captions, voice cloning, sound generation, runs locally on your own machine.

Edit by asking, not clicking

Connect Claude, Cursor, VS Code, Ollama or LM Studio to the built-in MCP server and say "cut the pauses, caption it and export for YouTube". Twenty-five tools drive the real app while you watch it happen in your window.

YouTube-ready in one pass

Auto captions from on-device Whisper, cloned-voice narration on your tag points, a 1280×720 thumbnail builder, and exports mastered to YouTube's −14 LUFS target with a quality report on every render.

Get VidHelm

Windows installer with FFmpeg bundled, nothing else to install. Or run it from source anywhere Node runs.

⬇ Download the betaStar on GitHub ★

Developers:

git clone https://github.com/RandoTechNerd/VidHelm
cd VidHelm && npm install && npm run dev
Licences, plainly. VidHelm's own code is MIT. It also ships other people's work, and the important one is FFmpeg: the bundled FFmpeg and FFprobe binaries are GPL-3.0-or-later, not MIT. VidHelm runs them as separate programs and stays MIT itself, but if you redistribute VidHelm you carry those obligations with it. Every package that ships inside the app is listed in THIRD-PARTY-NOTICES.md in the repo, and inside the app under ? then Credits & licences.

Thanks. A good stretch of what VidHelm can do exists because inventinside kept sending feedback, and is now a contributor. Got an idea? Open an issue. If VidHelm saves you time, a coffee keeps the updates coming: please put "VidHelm" in the comment so it lands against the right project, and add the feature you want next while you are there.