Ssakdduk
Automatic subtitles from speech
Recognises the speech and produces a timed caption draft. Edit the lines, pick a style, and burn them in. The recognition model runs on your own machine, so a video with someone's voice in it never has to leave it.
- Processed in
- Your browser
- Server upload
- None
- Formats
- MP4 · MOV
- Best under
- 20 min · 1080p
Drag & drop MP4/MOV or click to browse
Files stay in your browser — never uploaded
Recommended: under 20 min / 1080p — on iPhone prefer Most Compatible (MP4) — larger files may take several minutes
FAQ
Before you start
- How accurate is it?
- Good enough as a draft. Clear speech in a quiet room comes out well; names and jargon often do not. That is why the captions land in an editor rather than straight onto the video.
- Why the download the first time?
- Because the model runs on your machine instead of a server. "Faster" is about 39MB, "more accurate" about 73MB, and the browser keeps it afterwards.
- What if there is no speech?
- A clip with no audio is reported as such. If there is sound but no speech — music, room noise — the result is empty or nonsense, and typing the lines is quicker.
01
02
03