Record a voiceover over a screen recording, or redo the picture and keep the sound

Which tool in the “Segment” popover fixes what

Every tool here starts the same way: click a segment on the timeline, or hold Shift and drag across the exact range. The selection has to sit inside one edit segment: if it crosses a cut, the “Segment” button is disabled. Then open the “Segment” popover on the transport bar under the preview and choose what to replace. The table lists what each command changes and what it leaves as you recorded it.

How to add a voiceover to a screen recording on Mac

Take a 5-minute tutorial. At 2:10 you said “open the Export menu” while clicking Share, and the sentence runs to 2:22. The clicks are correct, so only the voice needs to be recorded again.

The same steps narrate a take you recorded without speaking, even one recorded with no microphone at all: the command then creates the voice track. Select the segment and talk over it. If the video has cuts, go one segment at a time, because a selection can’t cross a cut.

1. Check the input first. The voiceover is recorded from the system default microphone, so open System Settings > Sound > Input and select the microphone you used for the original take.
2. In the editor, hold Shift and drag across 2:10–2:22 on the timeline. Start and end between sentences, where you were silent.
3. Write the new sentence and read it aloud once against the clock. It has to fit in 12 seconds, because the picture won’t wait.
4. Open the “Segment” popover on the transport bar under the preview and choose “Record voiceover · keep video”.
5. Say the sentence. Recording stops by itself at the end of the range. If you stop it earlier, the rest of the range stays silent.
6. Play from 2:05 to 2:27. Listen for a change in level or room sound at 2:10 and 2:22. If the new voice is quieter than the rest, sit as close to the microphone as you did in the first take and record the range again.
7. Export the MP4. The picture and the system audio are the original take from start to finish.

How to re-record the video and keep the audio

The opposite case comes up when the product changes. Your onboarding video explains a settings page from 3:05 to 3:30, and last week the page got a new layout. The explanation is still accurate. Select 3:05–3:30, choose “Rerecord video · keep audio”, and perform the same actions on the new page. Recording stops by itself when the 25 seconds have been recorded. The result has the new screen picture under the original camera, microphone and system audio.

Timing is the hard part. The narration was spoken at the pace of the old clicks, so rehearse with the original playing and note the cues: “Notifications” is said at 3:12, so the click on Notifications has to land there.

• Use the same window size and browser zoom as the original take, so the picture matches at both ends of the range.
• The camera picture in the range stays the original one. Only the screen is recorded again.
• Put the page in the state it had at 3:05: same account, same scroll position.
• If the new layout needs more time than the old narration gives you, select a longer range and use “Rerecord” to perform and say it again. “Rerecord” stops when the selected duration has been recorded.

Voiceover, full retake or remove: how to choose

Choose the smallest replacement that makes the range correct.

• Right clicks, wrong words: “Record voiceover · keep video”.
• Right words, outdated or wrong picture: “Rerecord video · keep audio”.
• Both wrong, such as a wrong click that you also described: “Rerecord”, which replaces screen, camera, microphone and system audio for the range. Like every “Segment” command it needs a selection inside one edit segment, and it stops by itself when the selected duration has been recorded.
• A cough or an aside over a picture that should stay: “Remove voice in selection”.
• A whole segment that should play without any sound: “Mute segment”, which silences both the microphone and system audio.

What voiceover recording can’t do

The voiceover uses the system default microphone. There is no microphone picker in the command, so set the input in macOS before you start. Recording ends at the end of the range: a sentence that runs long is cut off there, and a longer explanation needs a wider selection. If you stop early, the rest of the range is silent. The range has to sit inside one edit segment; across a cut the “Segment” button is disabled. Only the microphone track is replaced, so system audio in the range plays as recorded, and burned-in captions are hidden over a range where the voice was replaced.

Qreata has no noise reduction and no volume normalization. A voiceover recorded in a different room will sound like a different room; the microphone track has one Volume control, from 0 to 200%, for the whole track. There is no music track either.

It works on recordings made in Qreata, because the app doesn’t import video files. Keep the project after you export: it stays editable, so a voiceover can be redone next month. Qreata runs on Apple Silicon Macs with macOS 14 or later.

Adding a voiceover with the free Mac tools

The built-in recorder (Shift-Command-5) records your microphone together with the screen, so the free way to get narration is to speak during the take. Afterwards, Apple’s list of QuickTime Player editing features is trim, split into clips, rearrange, rotate or flip, add a clip, and remove audio or video. Recording a new voice onto an existing movie is not on that list. On the free route that takes a second app: a video editor that imports the .mov and records narration.

To see the difference in 10 minutes, use the trial: record a 2-minute walkthrough, say one sentence wrong on purpose near 1:00, select that sentence and choose “Record voiceover · keep video”. Export and listen to the join on headphones. The trial is 7 days with every feature; the license is $49 once for 3 Macs, not a subscription.

• Apple: record the screen on your Mac
• Qreata app pricing