How to Use Your Voice: Pace, Volume and Stress
Everyone tells second-language speakers to slow down. The research says that advice is half right, and that the half nobody mentions matters more.
Ask any teacher how to use your voice when presenting and you will get one word back. Slower.
It is good advice, up to a point, and the point arrives sooner than you think. What follows is where that point is, and the three things your voice does that matter more than its speed.
OneHow fast is too fast
Roger Griffiths recorded the same texts at three controlled speeds and tested how much lower-level learners understood: 200 words per minute, 150, and 100.
At 200 words a minute, comprehension suffered. At 150 and at 100 it held up.
So the ceiling is real, and it is lower than an excited speaker’s natural rate. Nerves push almost everyone above their normal speed without their noticing, which is the actual reason the standard advice exists.
Be clear about what this study was, though. Griffiths measured listeners hearing recordings, not presenters in a room. It tells you what an audience can absorb, not what to do with your own mouth.
TwoSlower is not automatically safer
Murray Munro and Tracey Derwing looked at how speaking rate affects the way listeners judge second-language speech. They found a curved relationship, not a straight one.
A straight line would mean slower is always better. A curve means there is a best rate in the middle — and that speech rated worse for being too fast can also be rated worse for being too slow.
Speaking slowly is not free. Past a certain point it costs you the same thing that speaking too fast costs you.
This matters because “slow down” is the only voice advice most learners ever get, and taken to its conclusion it produces a careful, flat delivery that makes an audience work harder rather than less. Aim for the middle of the band, not the bottom. Somewhere between about 120 and 150 words a minute is a sensible target for a prepared talk in English.
ThreeThe “speak fast to sound credible” advice
You may have met the opposite claim. In 1976 Norman Miller and colleagues found that faster speakers were more persuasive and were judged more credible, and that result has been repeated in business books ever since.
Fifteen years later, Stephen Smith and David Shaffer found the condition it depends on, and it reverses the advice completely.
When the audience disagreed with the message, rapid speech helped: it left them less room to argue back in their heads, and persuasion went up. When the audience already agreed, the same rapid speech hurt: it stopped them building their own supporting thoughts, and the message landed with less force.
Which gives you something useful. In a friendly room, slow down and let people agree with you properly. In a room with doubts, keep moving. But remember that this literature is about persuasion, not comprehension — and comprehension is the thing you cannot afford to lose.
FourLet your voice go down at the end
Here is a small change with more evidence behind it than most delivery advice.
Thomas Vaughan-Johnston and colleagues tested falling versus rising intonation at the ends of sentences, across three experiments with hundreds of listeners each. Falling intonation made speakers sound significantly more confident. Every time.
The persuasion effect came with a condition attached: the benefit depended on how strong the arguments were. With strong arguments, falling intonation helped. With weak ones it did not help, and could work against the speaker. If that pattern feels familiar, you have met it before with question openings — anything that makes an audience listen harder will amplify whatever is actually there.
The practical version: end your sentences downward. Many learners raise the end of every statement, partly out of uncertainty and partly because some first languages do it routinely. In English it reads as a question, or as doubt. Record yourself saying three plain statements and listen only to the last word of each.
FiveVolume, honestly
This section carries no study, and we are telling you that rather than dressing up an opinion. Nothing was found at build time that cleanly measures presentation loudness. So take the next three lines as craft, not evidence:
- Aim your voice at the back row, not at the front. The people in front will be fine.
- Volume comes from breath, not from your throat. If your voice tires or goes rough, you are pushing from the wrong place.
- Ask one person beforehand to raise a hand if they cannot hear you. Then you can stop wondering, and wondering is what makes speakers quiet.
SixStress is the one that carries the meaning
Now the part that most public speaking advice never reaches, and it is the most important thing on this page.
English marks the important word in a sentence by stressing it — longer, louder, and on a higher note. Move that stress and you change what the sentence means, without changing a single word.
Laura Hahn tested what happens when it goes wrong. She played listeners three versions of the same talk: primary stress correctly placed, incorrectly placed, and missing altogether. With correct stress, listeners recalled significantly more of the content and rated the speaker significantly more favorably.
Read that second finding again. Stress placement did not only affect how much got through. It affected what the audience thought of the person speaking.
An honest detail: listeners also seemed to process the correct version more easily, but that particular result did not reach significance. The two that did are the two above, and they are enough.
One sentence, five meanings
Say each line out loud and hit the bold word harder, longer and slightly higher. The words never change. The meaning changes every time.
- I didn’t say she took the money
- I didn’t say she took the money. — Somebody else said it.
- I didn’t say she took the money. — I hinted at it.
- I didn’t say she took the money. — It was someone else.
- I didn’t say she took the money. — She borrowed it.
- I didn’t say she took the money. — She took something else.
- Now do it with a sentence from your own talk
- Write one line you will actually say. Underline the single word that carries your point.
- Say it five times, stressing a different word each time, and listen to the meaning move.
Why this drill and not a pronunciation exercise: you can do this at any level, today, with words you already know. It trains the thing Hahn measured — and unlike individual sounds, it does not take months.
SevenYour accent is not the problem
Munro and Derwing asked native-speaking listeners to do three separate things with recorded second-language speech: write down what they heard, rate how hard it was to understand, and rate how strong the accent was.
The three measures turned out to be related but partially independent. Speech could be rated heavily accented and still be understood almost perfectly.
Be careful with what that does and does not say. A strong accent was still correlated with lower comprehensibility. The finding is that a strong accent does not necessarily reduce how well you are understood — not that accent never matters at all.
But it is clear enough to change what you practice. Losing an accent takes years and may never happen. Putting the stress on the right word takes an afternoon, and Hahn showed it moves both comprehension and opinion. Spend your time there.
Five minutes with a recorder
Record ninety seconds of your talk, then listen to it seven times, once for each line below.
- Count your words for one minute. Under 120 or over 160, adjust and record again.
- Listen only to the final word of each sentence. Is your voice going down, or up?
- Pick any sentence and ask: which word was I stressing? Was it the right one?
- Listen for the flat stretch. Every recording has one. That is the part to rebuild.
- Play it at half volume from across the room. Can you still make out the words?
- Mark one sentence where you sped up. It is almost always the part you are least sure of.
- Re-record that one sentence only, three times, until it sits at the same speed as the rest.
The short version
That is how to use your voice when presenting, in five lines. Above roughly 200 words a minute, listeners start losing you. Slower is not automatically safer — the relationship is a curve, with a bottom as well as a top. Speed moves persuasion in opposite directions depending on whether the room agrees with you. Let your sentences fall at the end. And if you have one afternoon, spend it on which word you stress, not on how you pronounce it.
Then work on the silence between the words, which turns out to matter more than either.
Hear yourself change speed
Module 2 of Base Camp, our free B1 public speaking course, is a pace drill: the same four sentences at three different speeds, recorded, so you can hear which one is actually yours.
Where this comes from
Every claim on this page is drawn from published research, except the section on volume, which says on the page that it is craft rather than evidence.
Griffiths, R. (1990). Speech rate and NNS comprehension: A preliminary study in time-benefit analysis. Language Learning, 40(3), 311–336.
Munro, M. J. & Derwing, T. M. (2001). Modeling perceptions of the accentedness and comprehensibility of L2 speech: The role of speaking rate. Studies in Second Language Acquisition, 23(4).
Miller, N., Maruyama, G., Beaber, R. J. & Valone, K. (1976). Speed of speech and persuasion. Journal of Personality and Social Psychology, 34(4), 615–624 — and the study that found its limits: Smith, S. M. & Shaffer, D. R. (1991). Celerity and cajolery: Rapid speech may promote or inhibit persuasion through its impact on message elaboration. Personality and Social Psychology Bulletin, 17(6).
Vaughan-Johnston, T. I., Guyer, J. J., Fabrigar, L. R., Lamprinakos, G. & Briñol, P. (2024). Falling vocal intonation signals speaker confidence and conditionally boosts persuasion. Personality and Social Psychology Bulletin.
Hahn, L. D. (2004). Primary stress and intelligibility: Research to motivate the teaching of suprasegmentals. TESOL Quarterly, 38(2).
Munro, M. J. & Derwing, T. M. (1995). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 45(1), 73–97.