Audio editing in iMovie: what it can do and what it cannot

By default, audio is locked to its video clip. Right-click the clip and choose Detach Audio and it becomes its own track underneath, which you can move, trim and edit independently.

This is the step that unlocks most of what follows. Without it you cannot extend audio under a different shot, cut a picture while the sound continues, or trim one without the other.

Once detached, the audio can be dragged to a different position, which is how you fix lip sync or place a voiceover deliberately.

Every clip has a horizontal volume line across it. Drag it up or down to set the level for that clip. Drag the small circles at either end inwards to create a fade in or fade out.

For volume changes within a clip, hold Option and click on the line to add a keyframe, then drag it. Two keyframes either side of a section let you dip the music under a voice.

iMovie also has automatic ducking: select the clip that should stay loud, open the volume controls, and tick Lower volume of other clips. It is quick, it is crude, and it is often good enough.

There is also a Normalize Clip Volume button, which raises a clip towards a standard level. Useful on a quiet interview, and it works per clip rather than across the project.

iMovie has a Reduce background noise slider under the noise reduction and equalizer tab. It is a single control with no profile step, which makes it simple and blunt.

Use it sparingly. Past about 30 or 40 per cent it starts making voices sound hollow and underwater, and the damage is much more noticeable than the hiss it removed.

The equalizer next to it is a set of presets rather than a proper EQ. Voice Enhance is the useful one for speech and lifts intelligibility a little. There are no bands to adjust and no numbers.

That is the entire repair toolkit. No compressor, no gate, no de-esser, no way to remove a specific sound.

There is a hard limit of audio tracks, so layering a voiceover, music, ambience and sound effects runs out of room faster than you expect.

There is no way to see the waveform in real detail, which makes precise speech editing guesswork. You cannot zoom to the level where breaths are visible as distinct shapes.

And there is no compression, so an interview where one person is much louder than the other cannot really be evened out beyond per-clip volume.

Removing filler words means cutting the video at each one, which fragments the picture to fix the sound. That is the clearest sign you are using the wrong tool for the job.

The sequence that avoids all of the above: record the audio, clean and level it in an audio tool, export a finished track, then bring that into iMovie and cut the picture to it.

This is how most video is actually made. Audio is finished first because it is harder to fix and because the picture can be cut to fit it, not the other way round.

If the audio is already in a video file, detach it, export the project as audio only, fix it elsewhere, and bring it back. Clumsy, and still faster than fighting iMovie's tools.

Audacity does this for free. Hilite does it in the browser with noise and filler removal included, and we make it, so weigh that as you like.