Add Audio to Video

The video is not re-encoded

This is the part worth understanding, because it is where this differs from the tools that upload your file.

Adding audio does not require touching the video at all — the picture is untouched and only the audio track changes. So the video stream is copied straight through rather than decoded and re-compressed. Two consequences: the visual quality of the result is identical to your original, not a second-generation copy of it, and the process is fast, because the heavy work of re-encoding video never happens.

Upload-based tools generally re-encode, because their pipeline normalises everything that comes in. You usually cannot see the loss on one pass, but it is there, and it compounds if the clip goes through another tool afterwards. Here there is nothing to compound.

Replace or mix

Replace swaps the existing audio for your file. Right for footage whose original sound is useless — wind noise, room hum, a silent screen recording, a timelapse.

Mix keeps the original audio and lays the new track underneath it. This is the one you want for anything with speech in it: a talking-head clip with a music bed, a demo with a voiceover over the original UI sounds. Mix needs the video to actually have an audio track to mix with, so use Replace for a clip that is silent.

One thing worth knowing about mixing: it combines the two at their existing levels, so if your music is loud relative to the speech, the speech loses. If the balance comes out wrong, the fix is to lower the volume of the music file before adding it rather than to try again with the same file.

Looping, and the length trap

Music is usually shorter than footage. Loop audio if shorter is on by default and repeats the track until the video ends, which is almost always what you want.

Turn looping off and something surprising happens: the output ends when the audio does. A 20-second track on a 60-second video gives you a 20-second video. That is standard behaviour for this kind of operation rather than a mistake, but it catches people out, so it is worth knowing which way round it works. Leave looping on unless your audio is longer than the video.

The offset delays the start of the audio, for a video that should open on a few seconds of its own sound before the music comes in. It applies when looping is off — so if you want an offset in Replace mode, set looping to No and make sure your track is long enough to cover the rest of the video. In Mix mode the offset works alongside the original audio as you would expect.

A note on music you do not own

The technical part is easy; the licensing part is where people come unstuck. A commercial track added to a video you post publicly will usually be detected, and the result is a muted video, a blocked one, or the revenue going to the rights holder. That applies whatever tool put the music there.

For anything public, use music you are licensed to use: a library you subscribe to, properly licensed royalty-free tracks, or work released under a licence that permits it — checking the terms, since many require attribution. For a private video, a family clip or something internal, it does not arise.

Related tools

Video to MP3 pulls the audio out of another video when the track you want is already inside one. Trim Video cuts the clip to length first, which is easier than matching music to footage you are still going to shorten, and Video Compressor reduces the finished file.