IELTS Speaking: Pronunciation
This lesson tests whether you understand how Pronunciation is actually scored in IELTS Speaking, which features examiners listen for, and how to train them, rather than how to sound like a native speaker.
What this question looks like
Pronunciation is one of the four criteria used to mark all three parts of the Speaking test (Part 1 introduction and everyday topics, Part 2 the long turn with a cue card, and Part 3 the two-way discussion), alongside Fluency and Coherence, Lexical Resource, and Grammatical Range and Accuracy, each worth 25% of your Speaking band. Unlike the other three criteria, Pronunciation is not primarily about vocabulary or grammar choices, it is about how clearly and flexibly you produce sounds, stress, rhythm and intonation over a sustained stretch of speech. There is no separate "pronunciation task", it is assessed continuously as you answer every question across all three parts, using a public band descriptor that runs from Band 1 to Band 9. Crucially, the descriptors do not reward a "native-like" or particular national accent; a strong regional or non-native accent that remains clear and easy to follow can still score highly.
Step-by-step approach
- 1Learn what the examiner is actually listening for: the Pronunciation band descriptors reward a wide range of pronunciation features used flexibly (individual sounds, word stress, sentence stress, chunking, intonation, rhythm) and, above all, overall intelligibility with minimal strain on the listener. Band 7+ requires 'a range of pronunciation features' with 'flexibility', Band 9 requires this 'effortlessly'; low bands are defined by a limited range of features and frequent unintelligibility, not by having a foreign accent.
- 2Diagnose your own weak spots before you train blindly. Record yourself answering three or four Part 1 and Part 2 questions, then listen back (or get a teacher/partner to listen) and tick off specific problems: dropped word-final consonants (e.g. 'an' instead of 'and'), flat sentence stress where every word gets equal weight, rising intonation on statements that should fall, running all words together with no pausing at clause boundaries, or substituting sounds your first language lacks (e.g. /θ/ becoming /s/ or /t/). Target the two or three most frequent problems rather than trying to fix everything at once.
- 3Practise at the level of connected speech, not isolated words. Examiners assess you across full answers, so drill features inside sentences: word stress in multi-syllable words you actually use (comFORTable, neCESSary, phoTOGraphy), sentence stress on the content words that carry your meaning (nouns, main verbs, adjectives, negatives) while function words (a, the, was, to) are said quickly and reduced, and linking between words (turn_it_off, an_apple) so speech flows in natural chunks instead of staccato single words.
- 4Control intonation deliberately to signal meaning: a falling tone to sound certain and finish a point, a rising tone for genuine yes/no questions or to show a list is incomplete, and pitch variation across a long turn (Part 2) to keep the examiner engaged and to mark where one idea ends and the next begins. Also build in short, natural pauses at grammatical boundaries (after a clause, before 'and', 'but', 'because') rather than pausing mid-phrase, since well-placed pausing is itself scored as good chunking.
- 5Shadow authentic spoken English daily for rhythm and stress, not accent imitation. Choose short clips (30 to 60 seconds) of clear spoken English, for example news interviews or podcasts, listen once for meaning, then speak along with the recording matching its stress and rhythm exactly, repeating three or four times. This trains your ear and mouth together far faster than reading pronunciation rules alone.
- 6In the real test, prioritise clarity and pace over speed or accent. Speak at a natural, unhurried pace so consonants and word endings are not swallowed, self-correct calmly if you mishear yourself mispronounce a key word rather than freezing, and remember that one imperfect sound in an otherwise clear, well-stressed answer will not sink your score, examiners judge the overall pattern across the whole interview.
Worked example
Describe a favourite song or piece of music. You should say: what the song or piece of music is when you first heard it what it is about and explain why it is your favourite.
The piece I keep coming back to is a jazz track called 'Blue in Green', originally recorded by Miles Davis. I first heard it during my university years, quite late at night actually, when a flatmate had it playing softly in the kitchen. I remember standing completely still, just listening, because I'd never heard anything so unhurried before. It doesn't really have lyrics, so it isn't 'about' something in a literal sense, but to me it captures a feeling of quiet reflection, almost like the sound of thinking. The piano and trumpet seem to have a conversation, each waiting patiently for the other to finish. What makes it my favourite is partly its restraint. Nothing about it tries too hard; there's space between the notes, and that space feels intentional rather than empty. I also associate it with a specific memory, that late night kitchen moment, so hearing it now instantly transports me back there. Whenever I'm overwhelmed or need to slow my thoughts down, I put it on, and within a minute my breathing changes. It's not the kind of song you'd play at a party, but for me it's become almost a private ritual. Whenever people ask about music that's shaped me emotionally, this is always the first thing that comes to mind.
This answer shows the specific pronunciation features examiners listen for rather than an accent. Content words like 'unhurried', 'restraint' and 'ritual' carry natural stress, making the key ideas prominent against unstressed grammar words. Intonation falls clearly at the end of statements such as 'listening' and 'finish', signalling confidence and sentence completion, while it lifts slightly on the listed idea 'quiet reflection'. Connected speech appears in phrases like 'kind of' and 'shaped me', which link smoothly instead of being pronounced as separate, clipped words. Individual sounds that often trip candidates up, such as the consonant cluster in 'breathing' and the vowel in 'thought', are shaped clearly enough to stay intelligible even at natural speed. None of this requires native-like intonation patterns; it simply needs the stress, rhythm and linking to support meaning, which is exactly what the pronunciation band descriptors reward.
Try it yourself
Speak for one to two minutes on the cue card below. Before you start, plan quickly on paper: which words in each sentence you would naturally stress, where your pitch might rise or fall, and any sounds you personally find tricky. Record yourself, then listen back and check whether your stressed words actually stood out and whether your linking sounded smooth rather than choppy.
Describe a hobby or activity you enjoy. You should say: what the hobby or activity is how you started doing it how often you do it and explain why you enjoy it so much.
Common mistakes
- !Chasing a 'native accent' instead of clarity: candidates spend hours imitating British or American vowel sounds while ignoring word stress and intelligibility, which the band descriptors actually reward far more heavily than accent authenticity.
- !Speaking too fast to seem fluent: rushing causes word-final consonants and whole syllables to disappear (e.g. 'las' for 'last'), which damages intelligibility even though the candidate feels they sound confident.
- !Flat, equal stress on every word: this makes English sound monotone and mechanical, and can make it genuinely harder for the examiner to identify which words carry the key information in a long turn.
- !Pausing mid-phrase to think of vocabulary: pausing awkwardly inside a clause (e.g. 'I went to... the... shop') is scored differently from pausing naturally at clause boundaries, and frequent mid-phrase hesitation is read as poor chunking, not just a fluency issue.
- !Ignoring intonation on questions and lists: using only falling tone everywhere makes yes/no responses and lists of examples sound abrupt or unfinished, reducing the sense of flexible control examiners look for.
Quick quiz
1. According to the IELTS Pronunciation band descriptors, what mainly separates a Band 8-9 answer from a Band 5-6 answer?
2. A candidate stresses every word in a sentence with roughly equal weight and speed. What effect does this have on their Pronunciation score?
3. Why is shadowing (speaking along with a recording) an effective way to improve IELTS pronunciation?
4. A candidate pauses briefly after each clause ('I grew up in a small town, / and I moved to the city for university, / because there were more job opportunities there.') Is this good pronunciation practice?
Practise this in a real IELTS test
Take a free Speaking test with expert evaluation and apply the technique under exam conditions.
Take a free Speaking testIELTS Speaking: Pronunciation — FAQ
Do I need a British or American accent to get a high Pronunciation score in IELTS Speaking?
No. The band descriptors do not mention or require any specific accent, and examiners are trained to assess intelligibility and control of features like stress, rhythm and intonation, not accent origin. Many Band 8-9 candidates speak with a clear Indian, Nigerian, Spanish or other accent; what matters is that the examiner can follow you with no real effort.
Which matters more for my score, individual sounds or stress and intonation?
Stress, rhythm and intonation generally matter more, because they affect intelligibility across whole sentences and long turns, whereas a single mispronounced sound rarely blocks understanding on its own. That said, sounds that genuinely confuse meaning (for example consistently confusing 'ship' and 'sheep') can hurt intelligibility and should still be corrected.
How can I practise pronunciation effectively without a teacher?
Record short answers to real Part 1 and Part 2 questions, then listen back specifically for stress placement, word endings and pausing, rather than just overall impression. Combine this with daily shadowing of short clear audio clips (news, podcasts) and reading sentences aloud while marking stressed words in advance, which trains the ear and the mouth together over just a few weeks.