Before you practise any speaking task, it helps to know exactly what the scoring system listens for. This lesson walks through the official scoring traits, the Pronunciation and Fluency bands, and a self-rating sheet you'll reuse for the rest of the Speaking course.
Throughout the Speaking lessons, claims are labelled by how solid they are. Official means it comes from Pearson's published material (the PTE Core Score Guide and Test Tips). Practice tip means it's teaching experience: useful, but not a rule. Unverified means it circulates widely among test takers but has no source we can find. Always check the latest official guide, as Pearson updates these documents.
What each task scores
The most important point first: content is scored. Per Pearson's PTE Core Score Guide, the five speaking tasks don't all score the same traits.
| Task | Traits scored | How content is judged |
|---|---|---|
| Read Aloud | Content, Pronunciation, Fluency | Each replacement, omission or insertion of a word counts as one error |
| Repeat Sentence | Content, Pronunciation, Fluency | Content 0–3; 3 means every word repeated in the correct order |
| Describe Image | Content, Pronunciation, Fluency | Content 0–5; 5 means all elements, their relationships, possible development and a conclusion |
| Respond to a Situation | Appropriacy, Pronunciation, Fluency | Appropriacy 0–3 |
| Answer Short Question | Vocabulary only | 1 for an appropriate word, 0 for an inappropriate one |
So Read Aloud, Repeat Sentence and Describe Image score content alongside pronunciation and fluency. Respond to a Situation scores whether your answer fits the situation. Answer Short Question scores only the word you give: beautiful pronunciation on a wrong answer still earns 0.
These are three parallel lines, not one replacing another. A Read Aloud that sounds smooth but skips half the words can do well on Fluency and Pronunciation and still lose heavily on Content.
The Pronunciation and Fluency bands
Both traits run from 0 to 5 and are shared by Read Aloud, Repeat Sentence, Describe Image and Respond to a Situation. In PTE Core the top band is called Highly proficient. Here is a plain-English summary of the Score Guide descriptors. For drills on both traits, see Read Aloud basics: phrasing and pace and Repeat Sentence.
| Band | Pronunciation | Fluency |
|---|---|---|
| 5 Highly proficient | All sounds easily understood; natural linking and dropped sounds; stress correct in every word and sentence | Smooth rhythm; no hesitations, repetitions, false starts or simplifications |
| 4 Advanced | Clear and unambiguous; a few minor distortions; all words easily understandable | Acceptable rhythm; no more than one hesitation, repetition or false start |
| 3 Good | Most sounds correct; some consistent errors make a few words unclear | May be uneven with more than one hesitation, but no long pauses |
| 2 Intermediate | Some sounds consistently wrong; consonant clusters simplified; at least two thirds intelligible | Uneven; two or three hesitations; at most one long pause |
| 1 Intrusive / Limited | Many sounds wrong; about a third of words hard to follow | Irregular phrasing, many hesitations |
| 0 | More than half hard to understand | Slow, laboured, mostly isolated words |
Look closely at what changes between bands 2, 3 and 4 for Pronunciation. All three mention consonant problems; the difference is range. Band 2 describes sounds that are wrong everywhere, band 3 narrows it to a few consonants in certain positions, and band 4 requires that no word is hard to understand. Moving from 3 to 4 isn't about losing an accent. It's about fixing the few sounds you get wrong consistently, especially final consonants and consonant clusters.
Fluency is defined by events you can count: hesitations, repetitions, false starts and long pauses. The next lesson shows how to count them.
What the scoring engine measures
PTE Core speaking is scored automatically. Pearson's published description says the system recognises words and also locates and evaluates segments, syllables and phrases in your speech, then uses statistical models to assess the performance. Pearson's validation material for this technology lists features such as:
| Feature | What you can control |
|---|---|
| Latency of the response | Start speaking promptly on tasks with no preparation time |
| Rate of speech | Measure your own words per minute (next lesson) |
| Position and length of pauses | Pause only between sense groups, never inside them |
| Stress and form of words | Stress content words, reduce function words, keep final consonants |
| Sounds in context | Targeted work on the sounds you get wrong (Read Aloud Advanced) |
This is a list of aspects measured, not a full model, and Pearson publishes no weights. Any claim like "pauses cost X points" or "fluency is 40%" is unverified.
One practical consequence: on tasks with no preparation time (Repeat Sentence, Answer Short Question), a "calm second before speaking" isn't neutral. It goes straight into latency.
A practice tip for Read Aloud: say every content word (nouns, verbs, adjectives, adverbs, numbers) clearly, and let function words (articles, prepositions, conjunctions, pronouns) pass quickly and lightly. In "The elephant is the largest living land mammal", the words elephant, largest, living, land and mammal carry the stress; the and is are light. This affects Pronunciation and Fluency only. Light is not missing: skip "the" and that's an omission on Content.
Where to put your effort
Compare two ways of reading the same sentence:
- A: every word perfectly pronounced, but with several pauses.
- B: a noticeable accent, but connected and smooth.
Fluency band 4 allows only one hesitation, so repeated pauses drop you quickly to band 2. Pronunciation band 3 still tolerates some consistent consonant errors. So between these two traits, B usually costs less than A.
That trade-off does not extend to content. Missing words in Read Aloud, an incomplete Repeat Sentence, a Describe Image that skips key elements, or a Respond to a Situation that misses the point all lose marks that fluency can't make up. The order of work in this course is: say everything that needs to be said first, then remove the pauses, then sharpen the sounds.
Three common habits that cost marks
| Habit | What it sounds like | Which trait it affects |
|---|---|---|
| Dropping final consonants | -t, -d, -s at the end of words fade out | Pronunciation, and it blurs word boundaries |
| Flat stress | Every syllable gets the same weight | Pronunciation (word and sentence stress) |
| Broken phrasing | Pausing before prepositions to think of the next word | Fluency |
None of these is about vocabulary or grammar. They are sound habits, and the next two lessons work on them directly.
A reminder about silence. Per PTE Core Test Tips, the microphone closes after 3 seconds of silence, and you should start answering when the status box changes to "Recording". Silence has two costs: past 3 seconds the recording stops, and even shorter gaps still count toward latency and pauses. Don't treat "under 3 seconds" as safe.
A worked example: Read Aloud and Respond to a Situation
Read Aloud. Take this practice passage: "The elephant is the largest living land mammal. During evolution, its skeleton has greatly altered from the usual mammal, designed for two main reasons." Mark content words for stress and function words for light reading. That trains rhythm. But Content still counts every word: reading "elongated" as "elongate" is a replacement, and skipping "particularly" is an omission. Good rhythm with missing words and complete reading with stumbles are two different losses; neither covers for the other.
Respond to a Situation. Practice prompt: you are thirty minutes late for a doctor's appointment; explain to the receptionist and ask if the doctor could see you later today. You get 20 seconds to prepare and up to 40 seconds to speak.
I'm really sorry, I'm running about thirty minutes late for my appointment today. Would it be possible for the doctor to see me later this afternoon instead?
It apologises, explains and makes a polite request, and it covers both things the prompt asked for. The Score Guide's top Appropriacy band requires the social demands of the situation to be covered in an appropriate level of detail, so leaving out either part would keep you below band 3.
Self-rating sheet for your recordings
Record one Read Aloud on your phone (first take, no retakes). Then listen three separate times, one trait per listen. For new passages, use the Read Aloud practice questions.
Listen 1: Content. Count errors against the text: replacements (including "it is" read as "it's"), omissions (including a, the, of) and insertions. Pearson doesn't publish an error-to-score conversion, so just aim for zero.
Listen 2: Fluency. Count long pauses, hesitations (stretched "uh", stalls mid-phrase), repetitions and false starts. Then pick the first band whose conditions you fully meet:
- 5: all four counts are zero, and no final consonants dropped.
- 4: hesitations + repetitions + false starts total one or fewer; no long pauses.
- 3: more than one hesitation, but still no long pauses; most words come in phrases.
- 2: one long pause; two or three events; at least one smooth three-word run.
- 1 or 0: more events than that, or two or more long pauses.
Listen 3: Pronunciation. Listen a day later without the text, which is the only way to hear yourself as a stranger would. Count dropped final consonants, simplified clusters ("asked" said as "ass") and misplaced word stress.
- 4: no word is hard to make out; common words stressed correctly.
- 3: a few words unclear, always from the same few sounds.
- 2: some sounds wrong everywhere; clusters often simplified.
Your self-rating is not the machine's score and can't be converted into an overall score. Its job is to tell you what to practise next: lowest in Fluency, go to the next lesson; lowest in Pronunciation, go to Read Aloud Advanced.
Practice checks: rate the same recording again a day later. If both ratings land within one band on every trait, you're using the sheet reliably. Also try writing out, from memory, which traits each of the five tasks scores.