How to Remove Filler Words From an Interview

Why interviews are full of filler

People stall when they think. In an interview, someone is answering on the spot, so the ums, likes, and you-knows come thick, and they come more from a nervous first-time subject than from a practiced one. Left in, they make a sharp answer sound unsure. Taken out, the same answer lands.

The problem is volume. A ten-minute interview can carry a few hundred filler words. Finding each one by ear and nudging the playhead to trim it is the kind of task that eats an afternoon.

The manual way versus the transcript way

On a timeline you scrub, listen, zoom into the waveform, and cut around each um without clipping the words on either side. It works, and it is slow.

Editing by transcript flips it around. The interview becomes text, every word tied to its moment in the video. A filler word is just a word you delete, and the video closes the gap where it was. One pass across the whole transcript handles the bulk of them at once.

How to do it in Orator

Open the interview video editor and import the footage. It transcribes on your machine as the file loads, so nothing is uploaded. When the transcript appears, run the filler pass. Orator marks the ums, uhs, and the rest right in the text.

Accept the pass and every flagged word is cut, with the picture closing around each one. If it misses a straggler, highlight it in the transcript and delete it by hand; the manual cut behaves the same as an accepted one. The mechanics are the same ones described on the filler-word remover page.

Do not over-cut

A little hesitation is human. Strip every last pause and beat and the subject starts to sound like a machine gun. Clear the obvious fillers, then read the answer back. If it sounds natural, stop. The goal is a person who speaks clearly, not one who never breathes.

The privacy angle

Interview footage is often confidential until it airs. Because the file is read locally and transcribed in the browser, it never lands on a third-party server, which keeps a pre-broadcast or NDA-bound interview on your device the whole time. That is the same reason to keep a guest's podcast audio off the cloud.

What to do

For your next interview cut, transcribe it locally and clear the filler from the text before you touch anything else. Open the interview video editor and run the pass on a real clip to see how much cleaner the answers get.