We’ve all had the misfortune of having to cut off the end of a piece of dialog very abruptly in order to avoid an issue of some sort; there’s a clunk in the track, or the thought needs to be cut off for the sake of story, or someone else starts speaking over the first person, or any of a million other reasons.  Here are a few edit and mix tips to help resolve the issue.

Before getting too far into this post, I want to refer you to this great article on basics of dialog editing by Jay Rose.  It covers the basics of phonemes and cutting at their boundaries.  So start there.

Now back to these tips on solutions to abrupt dialog edits.  The first and most obvious option is to add a bit of reverb to the end of the clip.  This is almost always necessary, because one of the main things lost by abruptly cutting dialog short is the resultant sound of the space that the final syllable (or phoneme, really) causes.  So a little well-programmed reverb on the last syllable goes a long way towards solving the problem.  Often this is enough to sell an edit.  WARNING: be sure that the reverb is mono just like the dialog; it’s a subtle but goofy mistake to use a stereo or surround reverb to ring out the end of a dialog edit when the entirety of the prior line was all mono.  Try it both ways for yourself, and you’ll see that it just doesn’t feel right.

If reverb alone isn’t enough, another common trick is to just time-stretch the final syllable.  First break the last syllable (or rather phoneme) from the rest of its clip, then stretch it out.  You can simply use the TCE trim tool, but by default this renders a new clip.  To play around with results, you’ll want to undo the stretch each time and then re-try a different length so that you’re not piling time stretch on top of time stretch for a single clip.  That would quickly destroy the audio.

A better bet is to take advantage of elastic audio.

Instantiate the monophonic elastic audio engine plugin on the track, and then stretch with the TCE trim tool to your heart’s content.  (QUICK NOTE: The image above shows the polyphonic elastic audio engine.  This is because it also allows for pitch shifting, covered later in the article.  However, if you don’t need to make pitch adjustments, stick with the monophonic engine for better results.) Nothing is permanently rendered, so you can keep adjusting as necessary.  If you need to pass your edit along to a mixer and don’t want to be concerned with keeping the elastic audio plugin in place, simply remove it and choose to commit its changes.

If neither of these options is getting it done, the next bet is to comp takes.  That is to say, try and find the appropriate phoneme from elsewhere in the recording and edit it into place.  This is obviously a tricky proposition.  By far the biggest issue is trying to match the correct intonation.  The speaker needs to be in a matching pitch/register for this to work.  Usually this is only easily possible in scripted content with multiple takes of a given line to choose from.

However, sometimes if you find a piece that’s close but not perfect, a little pitch adjustment can take it over the finish line.  There are a few options for attacking this.  First, you can take advantage of the elastic audio engine again.  First enable the polyphonic engine on the track in question.  Then right-click the clip using the grabber tool and bring up its elastic properties window.  From here you can apply real-time pitch adjustments.  I advise not going up or down by more than one semitone, and usually only a half-semitone (50 cents).

Alternatively, you can apply a pitch shift from another software vendor’s plugin.  A few good options are Serato Pitch-n-Time, Sound Toys Little Alter Boy, or the excellent time-and-pitch engine inside iZotope RX 5 Advanced.  Again, keep the change minimal.  Beyond a semitone, pitch shifting is probably not going to be the answer.

Another small tip that can help sell an abrupt end of a dialog clip is to bring the level down slightly at the end.  Most sentences are naturally spoken more softly at the end than the beginning, so lowering the level is a natural way to arrive at the end of a dialog clip.  This is of course usually a mix decision, but in the edit you can use clip gain to apply this adjustment and simply inform the mixer that it was done.

A hugely important finishing detail is to be sure to add some fill/room tone at the very end of the dialog. This is necessary to help keep the room from feeling like it completely drops out, something that the clipped dialog already tends to do.  A well-placed bit of fill keeps the sound of the space alive, helping to avoid extra attention being drawn to the dialog edit.

Lastly a little bit of cloth foley or other movement sound subtly played along with the abrupt edit can help smooth it out and again make the space feel continuous and not chopped.  It essentially helps bridge the edit and can distract attention away from the cut.

Dialog editing is an art, and there are many more tips and tricks that can help in challenging edit situations.  But this is a good list to get you started. Happy editing!