Clean up a take instead of recording it again EN

Cut the silences, mute the filler words, trim the ends, then filter the noise.

You can fix most problems with a take in the editor instead of recording it again. Cut the dead air, mute the "um"s and "uh"s, trim the start and end, then filter the background noise. Doing the steps in this order keeps you from redoing work, because some edits clear the results of others.

Your original recording isn't changed while you edit. See Editing is non-destructive: getting the original back.

Before you start

You need to be the recording's owner, and the recording has to finish processing first. See Why editing, sharing and captions wait for processing.
If you want to mute filler words, generate captions first. Rekorder finds fillers in the transcript, so it can't find any without one. See Generate captions for a recording.

1. Cut the silences

Open the recording's video page and go to the Edit tab.
Choose Remove silences and filler words. This opens the editor on the Audio tab.
Choose Detect. Rekorder highlights every quiet gap on the track. The defaults are a threshold of -30 dB and a minimum length of 0.5 seconds.
Play the recording to check the gaps, then choose Remove.

This cuts the gaps from both the picture and the sound. If the result feels choppy, raise the minimum length and detect again. For what the two settings do, see Remove all the dead air automatically.

2. Mute the filler words

Still on the Audio tab, choose Mute filler words under Filler words. Rekorder mutes each hesitation and leaves the picture playing. It recognizes "um", "uh" and similar sounds in English, German, French and Romanian. It doesn't mute words like "like" or "you know". See Mute filler words for what counts as a filler and how to fix the result.

If you hear a filler that wasn't caught, or a cough or other noise you want gone, mute it by hand. See Silence a section without cutting the picture.

3. Trim the start and end

Go to the Edit tab. Play the recording, or click the ruler above the track, to put the playhead where you want the video to start.
Right-click the track and choose Split at playhead.
Click the segment before the split to select it, then press Delete.
Move the playhead to where you want the video to end and split again.
Select the segment after the split and press Delete.

The same steps cut a mistake out of the middle. See Cut out a mistake, and reorder what is left.

Splitting or deleting clears any silences you've highlighted but not yet removed. If you trim first, run Detect afterwards. If you've already chosen Remove, nothing is lost.

4. Render, then filter the noise

Choose Apply & render at the top of the page and confirm. Viewers only see your edits after you render. Rendering also clears the recording's transcript, summary and chapters, so generate them again afterwards. See Save a draft vs Apply & render and Why rendering clears captions, summary and chapters.

When the render finishes, apply the noise filter:

Open the Edit tab and choose Background noise filter.
Pick a level and choose Apply. Medium is the recommended level. Use High for a noisy recording, and Low if your voice starts to sound thin.

Apply the filter after you render, because rendering resets it to Off. The filter reduces steady sounds such as fans and keyboard clatter. It doesn't remove voices or sudden noises. See Remove background noise from a recording.

If you change your mind

Undo in the Edit tab, or Ctrl/⌘+Z, takes back the last action.
Restore original discards every edit and brings back the original video. It can't be undone.
To remove the noise filter, choose Off and then Apply.