Every feature is there because a creator asked for it. Nothing upsold, nothing hidden behind a paywall.
On-device AI
Whisper transcription runs on your machine. No upload, no queue, no cloud bill.
Word-level timing
Each word gets its own timestamp. Karaoke, per-word highlighting, precise sync. All free.
99+ languages
Auto-detect or pick a language. Works on whatever your subject speaks.
Six caption presets
Highlight, Karaoke, Bounce, Reveal, Script, Chunky. Each is a fully-editable starting point.
Runs offline
Plane, cafe, bunker. If your machine has power, CapForge works. No network required.
Transparent export
Bake captions into your video, or export a transparent MOV to composite anywhere.
Pricing
It's free. Forever.
CapForge is independent software. No subscription, no tiers, no limits on how many captions you forge. If it saves you hours, consider buying me a coffee on Ko-fi.
Everything you'd reasonably wonder before downloading.
In short
CapForge is a free AI caption editor for macOS and Windows. Transcription runs on-device with a local Whisper model, so your video never leaves your machine. It needs no account and has no subscription — the app is free forever. Downloads are served from GitHub Releases.
Is CapForge really free?
Yes. No trial, no subscription, no feature paywall. It's a passion project. If it saves you time you can support it on Ko-fi, but you never have to.
Does it send my video to a server?
No. All transcription happens locally using an on-device Whisper model. Your clip never leaves your machine. You can airplane-mode the whole process.
Which languages are supported?
99+ languages from the Whisper family, including English, Spanish, French, German, Portuguese, Italian, Dutch, Russian, Chinese, Japanese, Korean, Arabic, Hindi and more. Auto-detect is on by default.
What export formats do you support?
Video: MP4, MOV, and transparent MOV (ProRes 4444 alpha). Subtitles: SRT, VTT, and plain text. Encoding is hardware-accelerated where available.
Is it open source?
The application is distributed freely and the source lives on GitHub. See the repository for license details and release history.
System requirements?
macOS 12 Monterey or later on Apple Silicon, or Windows 10/11 64-bit. Transcription is fastest with at least 8 GB of RAM and a modern CPU (Apple Silicon runs it very efficiently).
Does CapForge work offline?
Yes. Transcription uses an on-device Whisper model and runs entirely on your computer — you can put the machine in airplane mode and still caption a clip. Your video is never uploaded.
How big is the download?
The macOS build is about 156 MB and the Windows build is about 151 MB. Both are downloaded from GitHub Releases.
Where do I report bugs or request features?
On GitHub. Open an issue at github.com/FRSname/CapForge and describe the bug or the feature you'd like — it's a one-person project, so that's the fastest way to reach me.