[e925]
░ armory · video-clipping · compare

ffmpeg-static vs LR-ASD

Both in the video & clipping category. Side-by-side — pick the one that fits your stack tonight.

ffmpeg-static★★★★
✓ loya-tested🆓 free🔌 drop-in

Bundled ffmpeg that just works — no 'your version is missing libfoo' errors.

rating
4
tested
✓ loya-tested
cost
free
install
drop-in
stars
1,350
updated
5w ago
#ffmpeg#video#node#npm#subtitles#drawtext
avoid if

You only need basic trim/extract operations — system ffmpeg is fine for those.

open the full entry →
LR-ASD★★★★
🆓 free🐍 sidecar

The 2025 state-of-the-art for 'which face is actually talking.' Fast, tiny, accurate.

rating
4
tested
cost
free
install
sidecar
stars
109
updated
1y ago
#video#active-speaker#python#research#lightweight#open-source
avoid if

You're not building a pipeline yourself. This is a research model, not a product.

open the full entry →

why it matters · ffmpeg-static

If you've tried to use ffmpeg in a Node project, you've probably hit the 'no such filter: subtitles' or 'no such filter: drawtext' error — because whichever ffmpeg your machine has was compiled without the library you need. ffmpeg-static is an npm package that ships a full ffmpeg binary with everything common pre-enabled (libass, libfreetype, the full encoder set). Drop it in, point your spawn calls at it, stop fighting the system ffmpeg. Works on Mac/Linux/Windows. 50MB dep, scoped per-project.

why it matters · LR-ASD

LR-ASD is the newest open-source active speaker detection model (Springer IJCV 2025 paper). It tells your video pipeline which person in a multi-face frame is actually talking. Accuracy beats the older TalkNet approach and it's 23 times lighter — fast enough to run on every frame, not just samples. If you're building your own clipping or auto-crop pipeline and accuracy matters more than a pre-built library, this is the one to drop in. MIT, free, Python.

more video & clipping to compare

derived live from the armory manifest · same-category only