The 8 CELPIP Speaking tasks, explained
What each of the 8 CELPIP Speaking tasks asks, how the prep-and-speak timing works, and a simple structure for the ones people find hardest.
⚡ The short version
- Speaking is 8 short tasks in about 15–20 minutes, each with a few seconds to prepare.
- Almost every task fits one of two shapes: give an opinion, or describe something.
- Tasks 3, 4 and 8 are image-based — narrate what you see, clearly and fully.
CELPIP Speaking is 8 short tasks in about 15–20 minutes, each recorded into a headset with a few seconds to prepare and a tight window to talk. They look varied, but once you see the pattern, they collapse into just a couple of shapes you can practise.
How Speaking works
Each task gives you a short prompt, a brief prep window, then a fixed speaking window. There's no examiner in the room — you speak to a microphone while a timer runs. The whole section is over in under 20 minutes, which is exactly why the clock is the real test.
The 8 tasks at a glance
| # | Task | What it asks |
|---|---|---|
| 1 | Giving Advice | Advise someone in an everyday situation |
| 2 | Talking about a Personal Experience | Tell a short story from your life |
| 3 | Describing a Scene | Describe what's happening in an image |
| 4 | Making Predictions | Predict what happens next in that image |
| 5 | Comparing and Persuading | Compare two options and persuade someone |
| 6 | Dealing with a Difficult Situation | Handle an awkward situation tactfully |
| 7 | Expressing Opinions | State and defend an opinion |
| 8 | Describing an Unusual Situation | Describe an unusual image to someone |
Two shapes cover almost everything
- The opinion shape (Tasks 1, 5, 6, 7): answer → two reasons → an example → a quick close. Lead with your position so the structure carries you.
- The describe shape (Tasks 2, 3, 4, 8): overview → a few specific details → a reaction or prediction. Move left-to-right or top-to-bottom so you don't miss anything.
Pick the shape during prep, then just fill it in as you speak.
The image tasks (3, 4 and 8)
Three tasks put a picture on screen. The trap is staring at one corner. Instead, cover the whole image: who's there, what they're doing, where, and what mood it has. For Task 4 you describe the same scene but predict what happens next; for Task 8 you're describing an unusual picture to someone who can't see it, so be concrete and orderly.
The clock, not your English, is what trips people
Most people who underperform here understand the prompt fine — they freeze in prep or trail off at the end. A fixed routine fixes both; we walk through it in beating the speaking timer.
Ready to run all eight under the real timers? Practise Speaking and get AI feedback on structure, vocabulary and fluency.
