Templates
Summarize Spoken Text frames for PTE Core
A recording of a minute or so, played once, then eight minutes to write 20 to 30 words on what the speaker said. The limit is tight: there is room for the topic, the main point and one supporting detail, and nothing else. These frames hold that shape so your notes, not a template, fill the words.
Pearson scores content 0 when the summary does not reflect what the speaker said, and form 0 when it is outside 20 to 30 words. Either zero makes the item 0. With only thirty words available, a memorised opener such as "The speaker was talking about an interesting and important topic" uses a third of them on nothing and is recognised for what it is.
The thirty-word budget
Thirty words is two short sentences or one medium one. The scorer wants the topic, the main point the speaker made about it, and at least one detail that shows you followed the argument. The scoring traits are content, form, grammar, vocabulary and spelling, so the sentences must be complete and spelled correctly; Summarize Spoken Text feeds your Writing score as well as Listening.
| Slot | Words | What goes in it |
|---|---|---|
| Topic + main point | 12–16 | Who or what, and the central claim |
| Supporting detail | 8–12 | One reason, example or number from the talk |
| Spare | 2–4 | A connector, or a second detail if short |
Frames
Two-sentence frames
- "The speaker explained that ___. She noted that ___." (24 to 28 words when filled)
- "The talk was about ___, which the speaker said ___. One example was ___."
One-sentence frames
- "The speaker argued that ___ because ___, pointing to ___ as evidence."
- "According to the speaker, ___, although ___."
Use "the speaker" rather than guessing a name or a title; the scorer does not need one and a wrong guess is a content error. Choose a reporting verb that matches what you heard: explained, argued, described, warned, recommended. Each adds a little content for one word.
One sentence or two?
Two short sentences are safer. A single thirty-word sentence needs a subordinate clause and a comma or two, and a slip in either is a grammar loss; two sentences of thirteen and fourteen words each have one subject and one verb, which is hard to get wrong under time pressure. Use the one-sentence frames when the talk has a clear cause-and-effect or contrast that a connector captures neatly; otherwise, write two. The scorer does not reward complexity for its own sake, and the form rule counts words, not clauses.
A filled example
"Most people assume that the best way to remember something is to read it again. In our studies, students who re-read a chapter three times felt confident but scored no better a week later than students who read it once. The group that did best was the one that closed the book and tried to write down what they remembered, then checked. We call this retrieval practice. It feels harder and slower, and students rate it as less effective, but in every trial it produced better recall a week later, by a margin of around twenty percent. The lesson for anyone studying for an exam is simple: test yourself more and re-read less."
The speaker explained that testing yourself on material improves recall far more than re-reading it. She noted that retrieval practice feels harder but raised scores by about twenty percent.
Topic and main point in sentence one: self-testing beats re-reading. Detail in sentence two: it feels harder, and the twenty percent figure. Twenty-seven words, two full stops, two finite verbs, nothing spelled by guesswork. The frames contribute eight words; the rest came from the notes.
Taking notes while it plays
The audio plays once, so the notes matter more than in any other writing task. Write while listening, not after.
- First ten seconds: the topic. Write the noun the speaker keeps returning to.
- Listen for the claim. Phrases like "the point is", "what we found", "the lesson is" mark it. Write three words.
- Catch one number or example. Numbers are easy to hold and are strong content. Write it as a figure.
- Note the final sentence. Speakers usually restate the main point at the end. If your claim note and the ending disagree, trust the ending.
- Do not write full sentences during the audio. Six to ten words of notes is plenty; sentence-writing while listening means missing the second half.
Eight minutes, allocated
- Minute 1: read your notes and decide the main point in your head.
- Minutes 2–4: write the two sentences using one of the frames.
- Minute 5: count words. The screen shows the count; 20 to 30 is a hard rule. Trim a connector or add one detail to land between 24 and 28.
- Minutes 6–7: check spelling word by word, and check that each sentence has a subject and a finite verb.
- Minute 8: leave it. Rewriting at the last moment is how word counts drift out of range.
The mistakes that cost the most
- Nineteen or thirty-one words. Form is scored and the range is exact. Aim for the middle.
- A list of details with no main point. "The speaker mentioned students, re-reading, a chapter and twenty percent" is content without comprehension.
- Adding what you think. The task is what the speaker said. Your view of retrieval practice is not content.
- Spelling errors. A separate trait and the most common avoidable loss. If unsure of a word, use a simpler one you can spell.
- A template first sentence. "In this lecture the speaker discussed a very important topic" is ten of your thirty words gone before any content appears.
Practise with real recordings and the eight-minute clock on the task page. The scorer there reports form and spelling separately from content, so you can see whether the loss is in what you wrote or in how you wrote it.