Blog · Why Do People Ask Me to Repeat Myself? It's Not Accent
Resources
Most people blame their accent when they're asked to repeat themselves. A licensed speech-language pathologist explains why it's usually listener effort and fixable habits instead.
.jpg)
If you have ever finished a sentence and watched someone lean in with that small apologetic "sorry, what was that?", you know how quickly it chips away at your confidence. It happens in meetings, at coffee counters, on phone calls, in the middle of a story you were enjoying telling. And most people land on the same explanation: it must be my accent. I hear this constantly from the professionals I work with, and I want to offer a different, more accurate, and frankly more useful answer.
Most of the time, people ask you to repeat yourself not because of your accent, but because something in the delivery made you harder to process. Speech researchers separate whether you can technically be understood from how much mental effort it takes to understand you. It is usually that effort, driven by fixable habits like fading volume, fast pace, or unclear articulation, that triggers the "sorry, what?"
I am Angela Chung, a licensed speech-language pathologist (SLP), and I work with adult professionals on exactly this problem. Not children with speech disorders, not stage actors, but working adults who are competent and articulate in their heads and yet keep getting interrupted with requests to say it again. What I have learned over years of this work is that the accent almost always gets blamed for things the accent is not doing.
Let me give you the honest version. When a listener asks you to repeat something, one of two things has usually happened. Either the signal genuinely did not arrive intact, meaning some sounds got dropped or blurred and there was not enough information for their brain to reconstruct the word, or the signal arrived but it cost them too much effort to decode in real time, so they gave up and asked rather than guess wrong.
That second scenario is the one people underestimate. Speech scientists talk about a difference between intelligibility and listening effort. Intelligibility is the technical question: if we wrote down what the listener heard, would it match what you said? Listening effort is a separate thing entirely. It is how much mental work the listener had to spend to get there. Two people can be equally intelligible on paper and yet one of them is exhausting to listen to and the other is easy. The exhausting one gets asked to repeat far more often, not because they are less clear in some absolute sense, but because listeners have a limited budget of attention and yours ran the meter up.
This matters because it reframes the whole problem. You are not trying to eliminate an accent. You are trying to lower the effort it takes someone to receive you.
Usually, no. Or at least, not in the way you think.
Here is a question I get often: does accent affect how people understand me? It can, but its role is smaller and more specific than most people assume. An accent is simply a consistent set of pronunciation patterns. Human brains are remarkably good at adjusting to consistent patterns. If you have ever spent ten minutes on the phone with someone whose accent was unfamiliar and found them perfectly clear by the end, you have felt this adaptation happen in real time. The accent did not change. The listener's brain tuned to it.
What actually creates trouble is inconsistency and low-information delivery. When sounds get swallowed, when volume collapses at the ends of your thoughts, when words fuse into each other so the listener cannot find the boundaries, there is nothing stable for their brain to lock onto. That is not an accent problem. A person with no discernible accent at all can be maddening to follow if they mumble the back half of every sentence.
So when someone asks me why does my accent make it hard to be understood, my honest answer is that the accent is often carrying the blame for habits sitting right next to it. Those habits are the good news, because habits are trainable in a way that your entire linguistic background is not, and does not need to be.
These are the patterns I see most, and the ones worth checking yourself for. You will probably recognize at least one.
This is the most common one by a wide margin. People start a thought at full strength and let it trail off, so the last three or four words, which are often the ones carrying the actual point, arrive at a whisper. The listener got your setup and lost your conclusion. They ask you to repeat the whole sentence, and you repeat the whole sentence, and you fade out again in the same place. The fix is not to be louder overall. It is to keep energy through the end of the thought, to treat the last word as if it matters as much as the first, because to your listener it usually matters more.
Fast speech is not automatically unclear, but fast speech gives the listener no recovery time. When they miss one word, and everyone misses words, a slower speaker gives them a beat to catch up and infer it from context. A fast speaker has already buried that missed word under six more. Speed also tends to compress your articulation, so the two problems compound. I rarely ask clients to speak dramatically slower. I ask them to add small pauses at the natural joints of a sentence, which does more for comprehension than an overall slowdown and feels less unnatural to do.
Clear consonants require physical movement, lips, tongue, jaw. Many of us, especially when tired or self-conscious, minimize that movement and speak from a nearly closed mouth. The vowels survive but the consonants blur, and consonants are what carry most of the distinguishing information between similar words. Under-articulating is one of those things that feels fine from the inside and lands as mush from the outside.
In natural speech we connect words, that is normal and you should not try to speak in robotic separated units. But there is a point where connection becomes collision, where "did you eat yet" is delivered as a single unbroken smear. When your listener cannot find the edges between words, they have to do the segmenting work themselves, and that is precisely the kind of extra effort that pushes them to ask you to repeat. Keeping the key words in a sentence just slightly more distinct gives the listener footholds.
English leans hard on emphasis. In any given sentence one or two words carry the weight, and a listener uses that stress to know what you actually mean. When everything is said at the same flat level, or when the stress lands on the wrong word, the listener gets the words but not the point, and "the point" is what they were listening for. This is subtle and it is one of the higher-leverage things I work on with clients, because fixing it changes comprehension without changing a single sound in your accent.
Sometimes it genuinely is, and it is worth being fair to yourself about this. Listening is not a one-way transaction. If the room is noisy, if the person is half-looking at their phone, if you are on a bad connection, the effort budget I mentioned earlier is already half spent before you open your mouth. Background noise and divided attention are real, established factors in how well speech gets received. On a crowded train platform, even a professional broadcaster gets asked to repeat themselves.
So no, you should not internalize every "sorry, what?" as evidence of a personal flaw. The useful question is about pattern. If it happens occasionally, in loud rooms, with distracted people, that is life. If it happens consistently, with attentive listeners, in quiet rooms, across different people, then the common factor is your delivery and that is worth working on. Not because anything is wrong with you, but because a few adjustable habits are standing between you and being received the way you intend.
This is the hard part to do alone, and I want to be straight about why. You cannot hear yourself the way a listener hears you. Your own brain fills in the words you meant to say, so the volume you dropped and the consonants you smudged sound complete to you. This is exactly why "just speak more clearly" is such useless advice. You are already speaking as clearly as you can perceive.
A few things help. Recording yourself, ideally in a real conversation rather than reading a script, and listening back specifically for where your energy fades or your words collide, gets you surprisingly far on your own. Ask a trusted colleague not "was that clear?" but the more specific "where did you lose me?" People can often point to the exact spot. And pay attention to the pattern of when you get asked to repeat, because the situation usually tells you the cause. If it is always the ends of your sentences, that is a volume and energy issue, not an accent issue.
Working with a speech-language pathologist adds the part you cannot easily do yourself, which is an outside ear trained to identify the specific mechanism and target it directly rather than telling you to try harder in general. When I describe what working with a coach actually looks like, this is the core of it: we find the two or three habits that are costing you the most listener effort, and we drill them until the new pattern feels normal, so it holds up when you are tired or nervous and not thinking about it.
Yes, and this is the part I most want people to hear. The goal is not to erase how you sound or to trade your voice for someone else's. Plenty of people with strong, obvious accents are effortless to listen to, and plenty of people with no accent at all are exhausting. The variable is not accent. It is clarity, energy, pacing, and emphasis.
The work I do in accent modification is really work on those variables. The name of the field is a bit of a historical accident, because for most of my clients we barely touch the accent itself. We make the delivery around it clearer, so the accent stops being the thing anyone notices and your ideas become the thing they receive. You get to keep sounding like yourself and simply become easier to follow.
If you take one thing from this, let it be that "sorry, what?" is information, not a verdict. It is telling you that something in the delivery cost your listener more than they had to spend, and almost always that something is specific, mechanical, and changeable. It is your volume falling off a cliff at the end of a thought. It is a pace that leaves no room to recover a missed word. It is consonants that never quite got their movement. None of those are your accent, and none of them are permanent.
So the next time it happens, resist the old reflex to blame the way you sound. Ask instead: where exactly did I lose them, and what was I doing right at that spot? That single question, asked honestly and often enough, tends to reveal the real pattern faster than years of quietly assuming it was the accent all along. What would it change for you to stop treating your voice as the problem and start treating your delivery as the thing you get to shape?
.png)
