Fix sound that is out of sync with the picture

Choose the video, say whether the sound is early or late and by how much, and download the fixed file. Nothing is re encoded.

Free, no sign upNo quality lossNothing is uploadedEarly or late sound

The sound is

How it works

  1. Step 1

    Choose the video

    Drop it on the tool or pick a file, up to 250 MB and 10 minutes.

  2. Step 2

    Say how far off it is

    Watch a clap or a hard consonant such as p or b. If you hear it before you see it, the sound is early; after, it is late.

  3. Step 3

    Download and check

    The picture and sound are copied, not re encoded, so it takes seconds. If it is still off, try the next step.

Offset or drift?

Check the first and last few seconds before choosing. This tool fixes an offset: the whole soundtrack moves by the amount you pick.

OffsetDrift
What you seeEarly or late by the same amount throughoutIn sync at the start, further out by the end
Fixed by this toolYesNo: convert to a constant frame rate or stretch the audio

How it works

FFmpeg reads the file twice and delays one copy with its itsoffset option: the sound when it is early, the picture when the sound is late. The two streams are then copied into one file without re encoding, so nothing loses quality.

Platform facts checked on 17 September 2026 against each platform's own help pages.

Use it from an AI assistant or a script

Every setting is a URL parameter, so an assistant can send a link with the options already chosen. The tool still runs on the person's own device.

ParameterValuesDefaultMeaning
shift-0.5, -0.3, -0.2, -0.1, 0.1, 0.2, 0.3, 0.50.2The sound is

Sound is 0.2 seconds early

https://percify.io/tool/fix-audio-video-sync?shift=0.2ffmpeg -i input.mp4 -itsoffset 0.2 -i input.mp4 -map 0:v -map 1:a -c copy -movflags +faststart output.mp4

Sound is 0.3 seconds late

https://percify.io/tool/fix-audio-video-sync?shift=-0.3ffmpeg -itsoffset 0.3 -i input.mp4 -i input.mp4 -map 0:v -map 1:a -c copy -movflags +faststart output.mp4

Every free tool and its parameters, as JSON: percify.io/tools.json

Making talking videos?

Audio driven lip sync generates the mouth from your audio, so the result starts in sync. 2 credits a second from a photo and an audio file.

Make a talking video

Free to start · no credit card

Frequently asked questions

Everything you need to know before you start.

If the sound is early or late by the same amount all the way through, shift it: choose the video, pick Early or Late and the amount, and download. If it drifts further out over time, that is a frame rate problem instead.

Watch a clap, a door closing or a p, b or m sound. If you hear it before you see it, the sound is early. If you see it first, the sound is late.

No. The picture and sound are copied, not re encoded; only their timing changes.

Usually the audio was trimmed, sped up or replaced after rendering. Rendering again from the final audio is the cleanest fix; for a constant offset, this tool shifts it.

No. It is processed on your device with FFmpeg running in WebAssembly.