How to stop saying um

You say um when your mouth arrives before your sentence does. The thought is still forming, and rather than let the gap sit there, you fill it. It is not a flaw in your speech. It is your brain refusing to leave a silence unattended.

Filler also does a job. It tells the person listening that you have not finished, that the turn is still yours. In a conversation that is genuinely useful. Alone at a microphone, with nobody waiting to interrupt, it has nothing left to do.

That is worth holding onto, because it reframes the whole problem. You are not trying to fix how you speak. You are trying to get comfortable with a pause that feels enormous to you and lasts half a second to everyone else.

Almost everyone who listens back to their first recording has the same reaction: the ums are everywhere and they are all anyone will hear. Then they send it to a friend, and the friend says nothing about it, because the friend was listening to the point being made.

You are hearing your own recording with full knowledge of what you meant to say. Every filler marks a place where you fell short of the sentence in your head. A listener has no such reference. They only have what arrived, and what arrived was an idea with some texture around it.

This matters practically. If you try to eliminate filler completely you will end up speaking carefully, and careful is worse than imperfect. The target is fewer, not none.

There is one technique that works and it is almost too simple to trust. When you feel an um coming, close your mouth and wait. The thought lands, you say it, and the recording has a clean pause where a filler would have been.

It will feel excruciating the first few times. A one second silence feels like five when you are the one inside it. Listen back and you will find it reads as consideration. Pauses are what confident speakers sound like.

There is a bonus if you are recording rather than speaking live. A clean pause is an edit point. An um is a thing you have to cut around. Every silence you leave yourself makes the edit faster.

The instinct is to write everything out. It rarely helps. Reading a full script either sounds read, or you look up, lose your place, and fill the gap while you find it again.

Filler clusters at transitions. You finish a point, you have not decided what comes next, and the um covers the decision. Fix the transitions and you fix most of it. Five bullet points in order will do more than five pages of prose.

Say the first line of each section out loud before you record. Not the whole thing, just the opener. Knowing exactly how you start removes the moment where the um usually lives.

Filler is one of the few things software genuinely fixes. Hilite detects ums, uhs and likes across a recording and removes them in one pass, and because you edit against the transcript you can see them as words on a page rather than hunting a waveform. An episode that felt unusable often needs five minutes.

Two honest limits. Remove every single one and the result can sound oddly airless, because natural speech has texture and stripping all of it leaves something slightly synthetic. And filler in the middle of a sentence sometimes sits on a breath, so cutting it too tightly clips the word beside it. Listen back to anything you cut aggressively.

The bigger limit is that editing does not change how you speak. It cleans the recording. If you want fewer ums in the first place, that comes from the pause in step three, and from reps.

People who worry about filler are usually people with something considered to say. The hesitation is thinking, out loud, in real time. That is not a defect to engineer away.

Practise the pause, prepare your transitions, clean up what is left in the edit, and then publish. The version of you that says um four times and means every word beats the version that says it none and sounds like a press release.