IELTS Speaking Pronunciation: Why Unclear Words Cap Your Band
What the official IELTS descriptors actually say about pronunciation and accent, why bands 5 and 7 have no description of their own, and how to find the words that cost you clarity.
Direct answer
Pronunciation is scored on how easily an examiner understands you, not on how close you sound to a native speaker. The official descriptors say the opposite of what most candidates fear: at band 9, “Accent has no effect on intelligibility.”
What holds most candidates at band 6 is the line about individual words, not their accent. Because a candidate must fully fit every positive feature of a band before they can score it, a handful of recurring unclear words can hold your pronunciation score down even when your rhythm and intonation are good.
How pronunciation is scored
IELTS Speaking is marked on four criteria: fluency and coherence, lexical resource, grammatical range and accuracy, and pronunciation. Pronunciation is one of four criteria in IELTS Speaking, each worth a quarter of the Speaking score, which makes it roughly a sixteenth of your overall band.
Two notes at the end of the official Speaking band descriptors change how you should prepare:
(i) A candidate must fully fit the positive features of the descriptor at a particular level.
(ii) A candidate will be rated on their average performance across all parts of the test.
The first note is the one that matters for pronunciation, and the rest of this article is mostly about its consequences. The second means the short exchanges in Part 1 are marked on the same scale as the long turn in Part 2; there is no warm-up section that does not count.
Does your accent affect your IELTS Speaking score?
No. The descriptors mention accent only to say that it stops mattering once you are easy to understand.
Band 9:
Can be effortlessly understood throughout.
Accent has no effect on intelligibility.
Band 8:
Can be easily understood throughout.
Accent has minimal effect on intelligibility.
Read the wording carefully. It is not “has minimal accent.” It is “accent has minimal effect on intelligibility.” A speaker with a strong first-language accent that a listener follows without effort fits that description. A speaker with a lighter accent who mispronounces the key nouns in their answer does not.
This is the same distinction we cover in intelligibility versus accent, and IELTS happens to state it in writing. The other boards take the same position without saying so: accent is named nowhere in the TOEFL Speaking scoring guide, the Cambridge assessment scales or the TOEIC Speaking descriptors. There is also no preferred variety anywhere in the descriptors. The criterion names intelligibility; it does not name an accent to aim for.
Why band 7 has no pronunciation description of its own
Here is the entire band 7 pronunciation descriptor:
Displays all the positive features of band 6, and some, but not all, of the positive features of band 8.
Band 5 is written the same way, bridging bands 4 and 6. The odd bands are not separate standards; they are the gap between the two even bands on either side.
Combine that with note (i) and you get the rule most candidates miss. Band 7 requires fully fitting band 6 first, so your ceiling does not move until the floor is clean. Adding some band 8 features earns you nothing while a band 6 feature is still missing.
So the floor is where the work is. Band 6 in full:
Uses a range of phonological features, but control is variable.
Chunking is generally appropriate, but rhythm may be affected by a lack of stress-timing and/or a rapid speech rate.
Some effective use of intonation and stress, but this is not sustained.
Individual words or phonemes may be mispronounced but this causes only occasional lack of clarity.
Can generally be understood throughout without much effort.
Note that band 6 already permits mispronounced words. So does band 8, which allows “occasional lapses.” The scale does not ask for a clean record. It counts frequency and cost: at band 4, words are “frequently mispronounced, causing lack of clarity,” with “patches of speech that cannot be understood”; at band 6, the same mispronunciations cause “only occasional lack of clarity.”
That is the difference between a 4 and a 6, and it is measured in words.
The four phonological features examiners listen for
The descriptors use the term phonological features, which candidates usually call pronunciation features. Four are named, and they are not equally easy to practise.
Chunking. Grouping words into units of meaning and pausing at the boundaries between them. Band 4 manages “some acceptable chunking”; band 6 is “generally appropriate.” This is a habit of phrasing rather than a set of sounds, and it improves quickly. Our guide to connected speech covers what happens inside a chunk.
Rhythm and stress-timing. Band 6 names two things that damage rhythm: a lack of stress-timing, and a rapid speech rate. That second one is worth reading twice, because many candidates speed up deliberately to sound fluent, and the descriptor treats that as a pronunciation problem. See rhythm in English.
Intonation and stress. Band 6 has “some effective use of intonation and stress, but this is not sustained.” Band 8 asks for the opposite: “Can sustain appropriate rhythm. Flexible use of stress and intonation across long utterances, despite occasional lapses.” The word doing the work is sustain: the two-minute Part 2 turn is where this criterion is really tested, not the one-sentence answers. Word stress is the place to start.
Individual words and phonemes. The only feature the descriptors track by count rather than by quality.
Three of those four are properties of how you deliver a sentence, and a week of deliberate practice moves them. The fourth is different: it is a list of specific words that come out wrong, and it can only be fixed word by word. It is also the one that keeps candidates at band 6, because a single recurring word is enough to stop you from fully fitting the level.
Find the words that cost you clarity
You cannot hear your own unclear words in real time, which is why candidates who practise for months without recording themselves often stall. Yes, you can improve IELTS pronunciation without a teacher, if you replace the teacher’s job with a recording and one narrow question: which words did a listener have to work to identify?
The routine:
- Record real answers, not drills. Three Part 1 questions about your work, your hometown, and your free time, plus one full Part 2 turn. Drills tell you what you can do when you are concentrating. Answers tell you what you do under load.
- Play it back once and mark the replays. Every time you have to rewind to identify one of your own words, write that word down. That is the list.
- Look for repeats across recordings. The words that appear in every recording are the ones tied to your own vocabulary: your job title, your city, your degree, the thing you do at weekends. Those are worth an hour each. A word you will say once is not.
- Fix in place, not in isolation. Word alone, then the phrase around it, then the full answer at normal speed. A word that only works when you say it slowly on its own has not been fixed.
The usual culprits, in rough order of how often they cost clarity:
- Voiceless /θ/ in think, three, month, through, replaced by /t/, /s/, or /f/.
- /v/ merging with /w/ or /b/ in very, develop, advantage, available.
- Short /ɪ/ against long /iː/ in live/leave, fill/feel, it/eat.
- English /ɹ/ in rural, career, research, variety.
- Consonant clusters at word ends: asked, texts, sixth, worlds. See consonant clusters.
Do not work through that list top to bottom. Work through your list, which is shorter and specific to the answers you are actually going to give.
Phonema is built for this loop. You type or paste the sentence you need to say, record it, and get feedback on which words came out clearly and which did not, so the list of words to fix comes from your own answers rather than a generic syllabus. Everything runs on the device, so your recordings do not leave your phone. There is more detail in how to check your English pronunciation.
A four-week pronunciation plan for IELTS
Week 1: record a baseline and list your unclear words
Record three Part 1 answers and one two-minute Part 2 turn. Play each recording back once and mark every word you have to replay before you can identify it. Write those words in a list. Do not fix anything yet.
Resist the urge to re-record until it sounds good. The point of week 1 is an honest list, and a polished third take will not give you one.
Week 2: fix the words, one at a time
Take the four or five words that appeared most often. Say each word alone until it is stable, then in the phrase it belongs to, then inside the full answer. A word is not fixed until it survives the full sentence at normal speed.
Four or five words sounds like a small target, but that is the number that moves the band 6 line about “occasional lack of clarity,” and it fits inside a week.
Week 3: slow the delivery and group the ideas
Re-record the same answers with pauses between ideas rather than between words. Aim for a rate that lets the endings of words survive. Rapid speech is named in the band 6 descriptor as something that damages rhythm.
A useful check: mark where you would put a comma if you were writing the answer, then pause there and nowhere else. Pauses at meaning boundaries read as control. Pauses in the middle of a phrase read as hesitation, and that costs you on fluency as well.
Week 4: sustain it across a two-minute turn
Record full Part 2 answers and check whether the control you built in weeks 2 and 3 survives to the end. Band 8 asks for flexible stress and intonation across long utterances, so the test is the last thirty seconds, not the first.
Most candidates lose rhythm and word endings in the final third of a long turn, when they are searching for content. Preparing more content for Part 2 is therefore also pronunciation practice.
Where pronunciation practice stops and exam practice starts
Pronunciation is roughly a sixteenth of your overall band, and no amount of clarity will rescue an answer that runs out after twenty seconds. The other three Speaking criteria, the shape of a Part 3 answer, the one minute of preparation before the cue card, and the experience of being interrupted by an examiner all need a different kind of practice.
For that part, my other app, OralPrep, simulates the full three-part IELTS Speaking test and scores answers against all four criteria. It is also useful for the question that pronunciation practice cannot answer: whether your answers are long enough and developed enough. Their guide to practising IELTS Speaking without a teacher is a reasonable place to start.
Phonema does the narrower job: the words themselves, one at a time, until they stop costing you clarity. Using either tool for the other’s job wastes preparation time you do not have.
Common questions
Does your accent affect your IELTS Speaking score? The descriptors mention accent only to say that it stops mattering once you are easy to understand.
How much of your IELTS score is pronunciation? Pronunciation is one of four criteria in IELTS Speaking, each worth a quarter of the Speaking score, which makes it roughly a sixteenth of your overall band.
Why is band 7 pronunciation hard to reach? Band 7 requires fully fitting band 6 first, so your ceiling does not move until the floor is clean.
Can you improve IELTS pronunciation without a teacher? Yes, if you replace the teacher’s job with a recording and one narrow question: which words did a listener have to work to identify?
Which accent should I use? Whichever one you already have. IELTS marks intelligibility, and switching accents a month before the test usually makes you less consistent, not more.
Band descriptors quoted from the official IELTS Speaking band descriptors; see also IELTS scoring in detail.