On-device transcription finds spoken PII in your footage and you mute the segment containing each one (the review shows every span and the total) — the picture is kept as-is (copied byte-for-byte from MP4 and MOV). Nothing is uploaded, and you get a masked transcript too.
On-device transcription finds spoken PII in your footage and you mute the segment containing each one (the review shows every span and the total) — the picture is kept as-is (copied byte-for-byte from MP4 and MOV). Nothing is uploaded, and you get a masked transcript too.
Redact spoken sensitive information in a video while the picture stays untouched. GhostVideo transcribes the audio track on your device with Whisper, detects PII in the transcript — phone numbers, emails, credit-card and IBAN numbers, and optionally names — and lists every finding with its timestamp. Review, uncheck what should stay audible, and apply: the transcript segment containing each finding — usually the whole sentence around it — is silenced (the review shows each muted span and the total), while the video stream is copied byte-for-byte. Nothing is uploaded, there is no watermark, and you can download a masked transcript alongside the redacted video.
Drop your video
Drag an MP4, MOV, MKV, or WebM file onto the page. It opens in the GhostVideo studio with Mute private speech pre-selected — nothing is sent anywhere.
Scan on your device
Whisper runs locally to transcribe the audio track, then the GhostRedact detectors find sensitive data in the transcript — numbers, emails, cards, IBANs, and optionally spoken names.
Review and mute
Uncheck anything that should stay audible, then apply. The segment containing each selected finding goes silent (the review shows each span and the total), the video stream is untouched, and you can download a masked transcript.
Does redacting re-encode my video?
No — the video stream is stream-copied byte-for-byte; only the audio track is re-encoded to apply the silence. The picture quality is exactly what you dropped.
Is the audio really muted, not just hidden?
Yes. Silence is written into the output's audio track with a real filter — the muted segments contain silence, with no way to recover them later. Muting works per transcript segment, so a finding mutes the sentence or phrase around it, not just the exact words.
What about a WebM source?
WebM containers can't carry the AAC audio this produces, so the redacted output is normalized to an MP4 container. If your WebM holds VP9 video, most browsers (Chrome, Firefox) play the result; Safari may not — convert to MP4 first if that matters to you.
What about privacy?
Transcription, detection, and muting all run in your browser. No video, audio, or transcript ever leaves your device.
Muting spoken details, covering faces and on-screen text, and stripping media metadata.
Every GhostVideo tool runs entirely in your browser. Your file is never uploaded — there's no upload endpoint to send it to.