IELTS Speaking Pronunciation: What Examiners Want

Short answer: Pronunciation is one of the four equally-weighted IELTS Speaking criteria, worth 25% of your band. It is scored on features — word and sentence stress, rhythm, intonation, and individual sounds — and above all on intelligibility, how easily a listener understands you.

It is not scored on your accent. A clear regional accent can reach Band 9.

No criterion generates more anxiety and more myths than pronunciation. Candidates arrive convinced they must sound British or American, spend months chasing a "neutral" accent, and miss the features that actually move the band. The examiner is not comparing you to a native speaker.

They are asking a simpler question: how easy are you to understand, and how well do you use the features of English that carry meaning?

This guide sets out what pronunciation is really scored on, dismantles the accent myth, explains stress, rhythm, intonation and sounds, defines intelligibility, and finishes with a worked example and a practice plan.

For how it fits with the other three criteria, see our overview of how IELTS Speaking is scored.

What pronunciation is scored on (features, not accent)

The pronunciation descriptor rewards control of the features of spoken English rather than conformity to any single accent. Those features fall into two groups.

The first is what linguists call the segmental level — individual sounds, the vowels and consonants that distinguish "ship" from "sheep" or "vine" from "wine". The second, and the one candidates neglect, is the suprasegmental level: the music of speech.

That means word stress (which syllable you emphasise), sentence stress (which words you make prominent), rhythm (the flow of stressed and unstressed syllables), and intonation (the rise and fall of your pitch).

Higher bands reward candidates who control both levels and who use these features to make meaning clearer.

Crucially, the descriptor does not ask you to eliminate your first-language accent. It asks whether accent features interfere with understanding.

A candidate can carry an obvious Indian, Nigerian, Spanish or Chinese accent and still score at the top, provided their stress, rhythm and sounds are controlled enough that a listener follows them effortlessly.

The official public band descriptors are published at IELTS.org — Speaking Band Descriptors, and reading them once will show you that the word "accent" is defined in terms of its effect on the listener, not its origin.

As with the other criteria, pronunciation runs on a band ladder built around effort. At the lower bands, mispronunciations are frequent and the listener has to strain throughout, or can follow only in patches.

At Band 6, you show some effective control of pronunciation features, but the control is not sustained and lapses reduce clarity here and there.

At Band 7 and above, you use a wide range of features flexibly, any first-language accent has minimal effect on understanding, and the listener follows you with ease from start to finish.

The climb is about the range and consistency of your control, not about erasing where you come from.

You do NOT need a native accent (myth-busting)

This is the single most damaging myth in Speaking preparation, so it deserves stating plainly: you do not need a British, American, Australian or any other "native" accent to score Band 9 on pronunciation.

IELTS is taken by millions of people across dozens of first languages, and the test is explicitly designed to be accent-fair. Examiners are trained on speakers from every linguistic background and are marking intelligibility and feature control, not how "local" you sound.

Chasing a native accent is not just unnecessary — it is usually counterproductive, because the effort of imitating unfamiliar vowel sounds tends to make your speech more hesitant and less natural, which quietly damages your Fluency and Coherence mark instead.

What the myth gets wrong is the direction of causation. It is true that many high-scoring speakers happen to sound quite "neutral", but that is a side effect of good stress and clear sounds, not the cause of their score. Keep your accent; make it intelligible.

The energy you might have spent flattening your vowels is far better spent on word stress and intonation, which carry more marks and improve faster. A distinctive, confident, clear accent is an asset in this test, not a liability.

There is a practical corollary worth internalising: the examiner is a trained listener, not a casual one.

Part of their job is to attune quickly to the speech of candidates from every linguistic background, which means they meet your accent halfway far more readily than a stranger on the street would.

Your task is not to sound like them; it is to give them clean, well-stressed input to work with. Fixate on clarity, and the question of accent takes care of itself.

Stress, rhythm, intonation, individual sounds

Word stress is where most intelligibility is won or lost. English words carry stress on a particular syllable, and putting it in the wrong place can make a word genuinely hard to recognise — "photograph", "photographer" and "photographic" each stress a different syllable.

Getting stress wrong on a key word forces the listener to pause and decode, which is exactly what lowers the mark.

Sentence stress works at the level of the whole utterance: fluent English emphasises the content words (nouns, main verbs, adjectives) and reduces the grammar words (articles, prepositions, auxiliaries) to a light, quick mumble. Speakers who give every word equal weight sound robotic and are harder to follow.

Rhythm follows directly from sentence stress — English is stress-timed, so stressed syllables fall at roughly regular intervals and everything in between speeds up to fit.

Intonation is the pitch movement that carries attitude and structure: a rising tone can signal a question or that you are not finished, a falling tone signals completion or certainty. Flat, monotone intonation not only sounds dull, it removes the cues a listener uses to parse your meaning.

Only after these should you worry about individual sounds. Sound errors matter when they change meaning or force the listener to work, but a slightly non-standard vowel that never causes confusion barely dents the score. Prioritise accordingly: stress and intonation first, problem sounds second.

A useful way to feel the difference is to notice what English does to unstressed vowels: it reduces them to a neutral "schwa" sound. The "a" in "about", the "o" in "today", the "e" in "the" before a consonant all collapse to the same soft "uh".

Speakers whose first language gives every vowel its full value tend to sound staccato and are paradoxically harder to follow, because the listener loses the rhythmic contrast between strong and weak syllables that English uses to chunk speech into meaning.

Deliberately weakening your grammar words is often the single fastest change you can make to sound more natural and be easier to parse.

Intelligibility is the goal

Every feature above serves one master criterion: intelligibility, meaning how easily and how consistently a listener understands you. The examiner is effectively asking, across the whole interview, "How much effort do I have to make to follow this person?"

At the lower bands, the listener has to concentrate hard and still loses parts of what you say. At the higher bands, understanding is effortless and sustained — not just in a slow, careful sentence, but throughout a natural-paced answer.

Consistency matters: a candidate who is clear when speaking slowly but becomes muddy at speed has an intelligibility ceiling that shows up the moment the topic gets demanding.

This reframes what "good pronunciation" means for your preparation.

You are not trying to impress the examiner with a beautiful accent; you are trying to remove the small frictions — a misplaced stress, a swallowed ending, a flat monotone — that make a listener work harder than they should.

Every friction you remove raises intelligibility, and intelligibility is the band.

This is also why pronunciation cannot be judged from one sentence: the examiner listens for whether you stay clear across the full range of what the test throws at you, which is why sustained practice beats last-minute polishing.

A worked example

The example below was written for this article as a teaching illustration, in the style of a Part 1 answer. The sentence is the same in both versions; only the delivery differs.

Sentence: "I usually take photographs of old buildings because the architecture is fascinating."

Lower-intelligibility delivery: every word is given equal, flat weight; "photographs" is stressed on the wrong syllable ("pho-to-GRAPHS"); word endings like the "s" on "buildings" disappear; and the pitch never moves. The listener can decode it, but has to concentrate, and "architecture" comes out with the stress misplaced so it takes a beat to recognise.

Higher-intelligibility delivery: the content words carry the beat — "usually", "PHO-tographs", "OLD BUILD-ings", "arch-i-TEC-ture", "FAS-cinating" — while the grammar words ("I", "take", "of", "the", "is") reduce and speed up. The word endings are audible, and the pitch falls naturally at the end to signal the sentence is complete.

Notice that nothing about the accent has changed; the speaker has not tried to sound British. They have simply put the stress in the right places, kept the endings, and let the intonation move.

That is the whole difference between a mark that says "generally intelligible" and one that says "effortless to understand".

How to improve pronunciation

The fastest returns come from stress and intonation, so start there rather than drilling individual vowels. Take new vocabulary and always learn where the stress falls — mark the stressed syllable when you record a word, and say it out loud with that beat exaggerated at first.

Our Word Coach is useful here because it feeds you words in natural context, so you can practise the stress pattern on real, high-value words rather than a random list.

For sentence stress and rhythm, read a short passage aloud and deliberately underline only the content words, hitting them and gliding over the rest; the shadowing technique — playing a short clip of clear English and speaking along a half-second behind — trains rhythm and intonation together.

Consistency is the trap most candidates fall into, so build practice around it. It is common to pronounce a word perfectly in isolation and then lose it inside a fast, complex sentence — which is exactly the situation the examiner cares about.

Rehearse your problem sounds and stress patterns inside full, connected speech at a normal pace, not just in single-word drills.

If a feature only holds up when you slow right down, it is not yet reliable, and the interview will expose it the moment the topic gets demanding and your attention shifts to content.

For individual sounds, be surgical rather than exhaustive.

Identify the two or three sounds from your first language that actually cause confusion — many Spanish speakers and "b/v", many Japanese speakers and "l/r", many Arabic speakers and "p/b" — and drill only those, because they are the ones that dent intelligibility.

Ignore sound differences that never cause a misunderstanding. Above all, record yourself and listen back, ideally comparing against a clear model, because you cannot fix a stress or intonation habit you cannot hear.

Sustained speaking practice is what consolidates all of this: the more you talk at length under realistic conditions — for instance rehearsing the Part 2 long turn with our cue card practice — the more your clearer patterns become automatic instead of something you have to think about.

If you want the criterion in its full context first, revisit how IELTS Speaking is scored.