How to use Video Redaction Editor
- Choose a supported MP4 or WebM. Enter a JSON list of rectangles using start/end seconds and x, y, width and height as percentages from the visible top-left corner.
- Set Preview time in seconds and select Preview masked frame. Check each important position and time boundary; rectangles do not track moving objects.
- Confirm both removal of all source audio and your mask review, then export with the tab visible. The new file is redacted-silent-video.webm.
- Reopen the downloaded WebM and inspect the whole sequence, especially mask entries, exits and movement. A preview or review checkbox does not establish that every sensitive frame was covered.
Example: Video Redaction Editor
Cover a fixed rectangle for the first two seconds of a three-second, 320 × 180 video.
Supported inputs and limits
- Up to 50 MiB, sixty seconds, 1920×1080 pixels and one hundred static rectangles with time ranges.
- Requires browser video-frame callbacks, manual canvas-frame capture and a WebM MediaRecorder encoder.
- The source is decoded and recorded in real time, so dropped or repeated frames, duration differences and quality loss are possible.
- Export stops on hidden-tab state, cancellation, timeout or a 100 MiB output limit.
- All audio is removed; selective audio redaction and moving-object tracking are not included.
More details
Source container metadata is not copied.
Review actual output frames rather than trusting the preview.
No anonymity guarantee is provided.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
Questions about Video Redaction Editor
Can I keep the original audio?
No. This initial mode removes the entire audio track to avoid retaining speech or other identifying sound.
Does a rectangle follow a moving face?
No. Enter additional static rectangles and time ranges, then review all transitions. Automatic tracking is not included.
Why are masks slightly longer than the entered range?
A 0.1 second margin at each boundary reduces exposure from browser frame timing. You must still inspect the recorded output.