IELTS speaking band 7 pronunciation

Here is the fact that changes how you should prepare, and almost nobody is told it: in the public IELTS Speaking band descriptors, Band 7 has no positive description of its own. Bands 4, 6 and 8 each describe concrete things a candidate does. Bands 3, 5 and 7 are defined as showing all the positive features of the band below plus some — but not all — of the positive features of the band above.

So a Band 7 in pronunciation is, precisely: everything a Band 6 does, done without gaps, plus a partial helping of what a Band 8 does. Nothing else. That is not a vague target, it is a two-part one, and the two parts require completely different work.

This is a paraphrase for planning purposes. The official public band descriptors are published by IELTS and are freely available — read them in the original before you rely on any summary, this one included.

Band 6 in pronunciation is broadly: you use a range of pronunciation features but your control of them is mixed, some effective use appears but is not sustained, and you can generally be understood throughout although mispronounced words or sounds reduce clarity at times.

The important word is generally. If there are moments where the examiner genuinely loses you, you are not at 6, and no amount of band-8 flourish elsewhere will pull you up. So the first job is a subtraction job: find the things that create those moments and remove them.

In practice there are only three common causes, in this order of frequency:

Wrong word stress. This is the biggest single one. A word stressed on the wrong syllable is often not recognised at all — the listener does not hear an accented version of the word, they hear no word. Stress errors on long, common academic words do more damage to a band than any vowel could.

One or two consistently substituted sounds. If you always produce th as s, or never distinguish /v/ from /w/, the effect accumulates across a whole test. Note that a single substitution is rarely fatal on its own — the problem is having three or four at once.

Mumbling or trailing off. Volume and final-consonant clarity are pronunciation features too, and they collapse exactly when candidates are unsure of their content. This is a nerves problem wearing a pronunciation costume, and it responds to rehearsal rather than phonetics.

Band 8 is broadly: a wide range of pronunciation features, sustained and flexible use of them with only occasional lapses, easy to understand throughout, and a first-language accent that has minimal effect on intelligibility. You do not need all of that for a 7 — you need some of it, visible and repeated.

Two words in there are doing the real work, and they are the two most candidates never address.

Range means you use more than one pronunciation feature. Not just accurate sounds — also stress, rhythm, intonation, pausing, emphasis for contrast. A candidate with flawless individual sounds and completely flat delivery is demonstrating a narrow range, which is a band-4 characteristic even if every phoneme is correct.

Sustained means it lasts. Band 6 explicitly includes effective use that is not sustained. Everyone can produce two good sentences. The difference between 6 and 7 is very often that the good version survives the whole of Part 2 rather than the first fifteen seconds, and then falls apart when the topic gets harder in Part 3.

This is why band-7 work is mostly stamina work, not knowledge work. You almost certainly already know how to do things you cannot yet keep doing.

Record two minutes of yourself answering a Part 2 style prompt. Then play it back six times, listening for exactly one thing each time. Listening for everything at once finds nothing — this is the single most useful change most candidates can make to their practice.

Whichever of the six is worst is your entire practice plan for the next fortnight. Do not spread effort evenly — the band is held down by the weakest feature, not raised by the strongest.

If you do one thing from this page, do this. Take the academic words you actually use in Speaking answers and confirm the stressed syllable of every single one. Play each, repeat it immediately at the same speed, and exaggerate the stressed syllable until it feels excessive. It will not sound excessive to a listener.

Note what happens to the unstressed syllables in these words: they shrink almost to nothing. That is the schwa doing its job, and producing it is half of what makes English rhythm recognisable. The schwa guide covers it properly, and English word stress covers the rules for predicting placement in words you have not met.

Your accent. The descriptors are built around intelligibility and range, not resemblance to a native speaker. Band 8 refers to a first-language accent having minimal effect on intelligibility — the accent is assumed to be present. Time spent trying to erase it is time not spent on the six features in section 4.

Which native accent you model. British, American, Australian, Irish, Indian — examiners are trained across varieties and none scores higher. Consistency matters more than choice, because mixing features from several accents is what sounds unstable.

Speed. Fast is not fluent. Speed with no chunking scores worse than a measured delivery with clear phrase boundaries, and it damages your pronunciation score by compressing exactly the features you are being marked on.

Vocabulary size. That is a different criterion. Using an ambitious word you cannot stress correctly costs you in pronunciation what it gains you in lexical resource, and frequently more.

Week 1 — subtract. Record, run the six-pass diagnostic, and list every word you mispronounced. Fix the stress on all of them. Nothing else this week.

Week 2 — rhythm. Take short paragraphs, mark the content words, and read them aloud giving only those words weight. Reduce everything else. The target is that a listener could tap along.

Week 3 — chunking and intonation. Same texts, now marked with slashes at phrase boundaries. Pause at the slashes and nowhere else. Add a deliberate pitch fall at the end of statements and a rise on list items.

Week 4 — stamina. Full two-minute answers on unfamiliar topics, recorded, with one rule: no restarts. You are testing whether weeks 1 to 3 survive under load. If control drops after forty seconds, that is the finding, and the fix is more two-minute answers, not more theory.

Speak slightly more slowly than feels natural, finish your consonants, and take the full preparation minute in Part 2 to note the difficult words you intend to use — not the ideas. You already have the ideas. What you need is thirty seconds of certainty about the stress on four or five words.

If you mispronounce something and notice, say it again correctly and move on. Self-correction is not penalised in pronunciation and the alternative — a silent flinch followed by thirty seconds of caution — costs more.

IELTS pronunciation words · Academic word list · TOEFL pronunciation words · Noun / verb stress pairs · Weak form words

Frequently mis-stressed in test answers: comfortable, opportunity, development, environment, particularly, vocabulary.

IELTS speaking pronunciation — start here for how the criterion works and what the examiner is listening for at any band. English word stress · The schwa · Connected speech · Minimal pairs practice — for finding which individual sounds are actually costing you clarity.

Read slowly first. Pronounce each syllable deliberately before trying to blend them at full speed. Rushing hides the vowel and consonant shapes your ear needs to learn.

Record yourself. Most phones have a voice memo app — record a word, then compare it to the playback here. Hearing the gap between your version and the model accent makes targeted practice far easier.

Focus on one accent. Jumping between American, British and Australian in the same session creates confusion. Pick the accent you encounter most and stick with it until the patterns feel natural.

Every word above is a button. Set the accent and voice you want on the main screen, then come back — the guides use whatever you chose. To hear two accents against each other, use Compare; to test whether you can actually hear a contrast, use the listening quiz.