CELPIP General
CELPIP Speaking Task 8: Describing an Unusual Situation | Complete Guide
CELPIP Speaking Task 8 at a Glance Aspect Detail Task Describing an Unusual Situation Prep time 30 seconds …
CELPIP Speaking Task 3 at a Glance
| Aspect | Detail |
|---|---|
| Task Name | Describing a Scene |
| Position | Task 3 of 8 in the CELPIP Speaking section |
| Prep Time | 30 seconds |
| Speaking Time | 60 seconds |
| Format | View an image on screen, then record into the microphone |
| Key Function | Describing, observing, speculating |
| Connected to | Task 4 (same image, predict what happens next) |
| Scoring | Content/Coherence, Vocabulary, Listenability, Task Fulfilment |
Task 3 is the first task in CELPIP Speaking that gives you a visual instead of a written scenario. You see an image. You describe what is happening. This is purely descriptive. No opinion needed. Just say what you see.
The visuals are realistic, ordinary scenes: a busy outdoor market, a family in a park, a coffee shop, a street, an office reception, a community gathering. Nothing unusual or surreal. That comes in Task 8.
What Is CELPIP Speaking Task 3?
In Task 3, an image appears on screen. You have 30 seconds to study it. Then you have 60 seconds to describe what you see.
You cannot pause the image. Once your 60 seconds are over, the test moves to Task 4. Importantly, Task 4 uses the same image. So Task 3 doubles as preparation for Task 4: anything you noticed but did not have time to mention can become a prediction in Task 4.
Tip
Use your 30-second prep window to count, not write. Count the people. Count the activities. Note 2 to 3 specific details in foreground, middle, background. That gives you a map for 60 seconds of speaking.
What Task 3 Is Not
- It is not a speculation task. Heavy guessing about backstories is the wrong function.
- It is not an opinion task. “I think this is a nice scene” wastes time.
- It is not a comparison task. Do not compare what you see to your own city.
- It is not a prediction task. That is Task 4.
You are a camera describing what is in the frame.
The Exact Format with a Worked Sample Prompt
You will see an image on screen with a short instruction line above it.
Sample CELPIP Task 3 Prompt
Describe what is happening in the picture. Mention everything you can see, including people, places, and activities.
(Image: A busy outdoor farmers’ market on a sunny afternoon. A woman in a red jacket examines fruit at a stall. Behind her, three people chat near a coffee stand. Wooden tables with vegetables and crafts. Colourful awnings overhead. Two children with a dog in the background.)
The image stays on screen during prep and during your 60-second recording. You do not need to memorise it. You can keep looking back at it as you speak.
How CELPIP Speaking Is Scored
Every speaking task is scored on the same four dimensions.
| Dimension | What It Means | What Scores Well in Task 3 |
|---|---|---|
| Content/Coherence | Did you address the task fully and logically? | Clear spatial organisation, 5 to 7 specific elements |
| Vocabulary | Range and accuracy of words used | Precise nouns, varied descriptive adjectives, prepositions of place |
| Listenability | How easy is it to follow your speech? | Steady descriptive pace, clear pronunciation of “-ing” forms |
| Task Fulfilment | Did you do what the task asked? | Stayed in description mode, no opinion or prediction |
Info
CELPIP uses human raters for Speaking, not AI. This means they can imagine the image you are describing and follow your spatial logic. They notice when you describe an image randomly versus systematically.
Scoring Bands for Task 3
| Dimension | Score 9-12 (CLB 9+) | Score 7-8 (CLB 7-8) | Score 5-6 (CLB 5-6) |
|---|---|---|---|
| Content/Coherence | Systematic spatial coverage. 5 to 7 specific elements with detail. | Most of the image covered. Some elements vague. | Random coverage. Repeats same element. Misses obvious features. |
| Vocabulary | Precise nouns, varied adjectives, spatial prepositions. | Adequate, occasional repetition of “nice” or “good”. | Limited range. “Person”, “thing”, “place” used repeatedly. |
| Listenability | Confident present continuous, clear pronunciation of “-ing” endings. | Generally clear, occasional tense slips. | Frequent grammar errors, listener works hard. |
| Task Fulfilment | Pure description. No opinions or predictions. | Mostly description, occasional drift. | Mixes description with opinion or speculation about why. |
Step-by-Step Approach: 30-Second Prep + 60-Second Speak
The 30-Second Prep Window
Do not write sentences. Scan the image strategically.
- Big picture (5 sec): What kind of scene is this? Where? Time of day? Mood?
- Foreground (8 sec): Who or what is closest? What are they doing? Count them.
- Middle ground (7 sec): Who or what is behind the foreground? Activities?
- Background (5 sec): What is in the distance? Sky, buildings, walls?
- Specifics (5 sec): Pick one or two precise details (colour of jacket, type of food, weather).
Warning
Do not try to identify everything in the image during prep. Even at CLB 10, 5 to 7 well-described elements beat a scattered list of 12. Depth beats breadth.
The 60-Second Speaking Window: Marvel’s Four-Beat Structure
| Section | Timing | What to Cover |
|---|---|---|
| Big picture first | 10 sec | One sentence framing the whole scene |
| Describe people | 20 sec | Foreground and middle ground people, what they are doing |
| Setting and objects | 15 sec | The space, the objects, colours, layout |
| Speculate or add detail | 15 sec | “It looks like a weekend market”, “Everyone seems relaxed” |
You should be at the 25-second mark when you finish describing people. That tells you to move to objects.
Verbatim CLB 10 Sample Response (with Annotations)
Here is a model response to the farmers’ market image described above.
Model Response: Farmers' Market Scene
(Big picture) This image shows a busy outdoor market on what appears to be a sunny afternoon. The atmosphere looks lively and relaxed.
(People) In the foreground, there is a woman in a red jacket examining some fresh fruit at a stall. She seems to be carefully choosing apples or pears. Behind her, a group of three people are chatting near a coffee stand, holding takeaway cups and smiling. Further back, two young children are playing with a small brown dog, while one of them is laughing.
(Setting and objects) The market has colourful awnings overhead in red, yellow, and green. There are wooden tables stretching across the scene, with baskets of vegetables, fresh bread, and handmade crafts on display. A wooden sign above one of the stalls says “Organic Produce”.
(Speculate and close) Overall, it looks like a weekend farmers’ market in a small town or community park. Everyone seems to be enjoying the warm weather, and the whole place has a really friendly, neighbourhood feel to it. The light suggests it is probably late morning or early afternoon.
Why This Response Scores High
- Content/Coherence: Clear spatial flow from foreground to middle to background to overall mood. 8 specific elements mentioned.
- Vocabulary: Strong adjectives (lively, relaxed, fresh, colourful), precise verbs (examining, chatting, playing, stretching), spatial prepositions (in the foreground, behind, further back, overhead).
- Listenability: Confident present continuous throughout. No tense slips. Pauses fall at sentence breaks.
- Task Fulfilment: Stays in description mode. The mild speculation at the end (“looks like a weekend farmers’ market”) is grounded in what is visible, not invented.
Sentence Stems and Opening Phrases
These are flexible patterns, not memorised lines.
Big Picture Openers
- “This image shows…”
- “In this picture, I can see…”
- “The scene takes place in…”
- “What I can see is…”
- “It looks like this picture was taken at…”
Describing People (Present Continuous)
- “In the foreground, a woman is…”
- “Behind her, two people are…”
- “On the left side of the image…”
- “Further back, you can see…”
- “There is a man who is…”
Describing Objects and Setting
- “There are several wooden tables…”
- “You can see colourful awnings…”
- “The background shows…”
- “Above the stalls, there is…”
- “Around the scene, I notice…”
Speculating and Closing
- “It looks like a…”
- “It appears to be…”
- “I would guess this is…”
- “Overall, the scene feels…”
- “The atmosphere seems…”
Tip
Spatial prepositions are gold in Task 3. “In the foreground”, “behind”, “further back”, “to the left”, “on the right”, “overhead”, “in the distance”. Use four or five different ones in a single 60-second response. It immediately raises your Vocabulary and Content scores.
Common Mistakes That Cost Marks
| Mistake | What It Sounds Like | Score Impact |
|---|---|---|
| Wrong tense | “The man walked…” or “The man walks…” | Listenability and Task Fulfilment drop |
| No spatial logic | Jumps from a person to a tree to a child randomly | Content/Coherence drops |
| Generic words | “Person”, “thing”, “place”, “stuff” repeated | Vocabulary drops |
| Predicting instead of describing | “I think they will buy fruit and go home” | Task Fulfilment drops sharply |
| Heavy opinion | “This is a nice picture and I like it” | Task Fulfilment drops |
| Repeating “I can see” | Used 8 times in 60 seconds | Vocabulary drops |
| Stopping at 35 seconds | Long silence at the end | Content/Coherence drops |
| Memorised opening | “Today I am going to describe a picture which shows…” | Listenability and Task Fulfilment drop |
Warning
The biggest Task 3 mistake: mixing present continuous with present simple inconsistently. If a man is mid-action, say “the man IS walking”. If something is permanent, say “there ARE tables”. Mixing them sounds careless. Pick the right form for each element.
CELPIP vs IELTS GT vs PTE Core Speaking
| Aspect | CELPIP General | IELTS General Training | PTE Core Speaking |
|---|---|---|---|
| Format | Recorded into microphone, 8 separate tasks | Face-to-face with examiner, 3 parts | Recorded into microphone, multiple task types |
| Visual description task | Task 3: 60 seconds describing an image | None directly. Some Part 3 questions touch on visuals abstractly. | Describe Image: ~25 sec prep, 40 sec speak |
| Prep time | 30 sec | None for IELTS Speaking | 25 sec for PTE Describe Image |
| Speak time | 60 sec | Varies | 40 sec for PTE Describe Image |
| Image type | Realistic scene with people | Not applicable | Often charts, graphs, diagrams |
| Examiner | None (recording only) | Yes, live examiner | None (AI scored) |
| Scoring criteria | Content, Vocabulary, Listenability, Task Fulfilment | Fluency, Lexical Resource, Grammar, Pronunciation | AI multi-trait |
Info
Key difference from PTE Describe Image: PTE images are usually data visualisations (line graphs, pie charts, process diagrams). CELPIP images are everyday human scenes (markets, parks, offices). PTE rewards numerical accuracy and trend vocabulary. CELPIP rewards spatial vocabulary, present continuous accuracy, and observational detail. Train differently.
How to Practise CELPIP Task 3 Properly
What Effective Practice Looks Like
- A bank of unfamiliar realistic scene images (parks, markets, offices, streets, cafes)
- A 30-second study timer
- A microphone that locks at 60 seconds
- Replay with image alongside (so you can review what you missed)
- Feedback on present continuous accuracy and spatial vocabulary range
A Weekly Practice Plan
| Day | Activity | Time |
|---|---|---|
| Mon | 4 fresh Task 3 images, recorded, no notes | 12 min |
| Tue | Listen back, mark tense slips, repeated words, missed elements | 15 min |
| Wed | Repeat Monday’s images using spatial preposition list | 12 min |
| Thu | 3 new images, focus on hitting 60 seconds with 6+ elements | 12 min |
| Fri | Pair Tasks 3 and 4 on the same image for prediction crossover | 20 min |
| Sat | Present continuous drill: 30 sentences in 5 minutes | 10 min |
| Sun | Full Speaking section practice | 25 min |
Three More Sample Scenes with Mini Plans
Sample Scene 2: Office Reception
Image: A modern office reception with a receptionist at a desk, two people waiting on a couch, plants near the window.
Quick plan:
- Big picture: a busy but professional office in the morning
- Foreground: receptionist at desk, typing
- Middle: two visitors waiting, one on phone
- Background: large window, plants, modern artwork
- Detail: glass walls, hardwood floor, name plate on desk
Sample Scene 3: Family Picnic in a Park
Image: A family of four on a picnic blanket, a dog nearby, trees, a couple jogging in the distance.
Quick plan:
- Big picture: a sunny weekend afternoon in a public park
- Foreground: family of four on a checked blanket, food spread out
- Middle: a brown dog rolling on the grass nearby
- Background: tall trees, a path, joggers in the distance
- Detail: the mother is pouring lemonade, a child is laughing
Sample Scene 4: Bus Stop in Winter
Image: A snowy bus stop with three people waiting, all bundled up, a bus approaching.
Quick plan:
- Big picture: a cold winter morning at a city bus stop
- Foreground: three people in heavy winter coats, hats, scarves
- Middle: a bus approaching from the right
- Background: snow-covered buildings, grey sky
- Detail: one person is checking their phone, another is holding a coffee
Vocabulary Bank for Task 3
| Category | Useful Words |
|---|---|
| People (verbs) | examining, chatting, browsing, holding, waiting, leaning, gesturing, laughing |
| Spatial prepositions | in the foreground, behind, further back, overhead, in the distance, on the left |
| Weather/light | sunny, overcast, snowy, dim, bright, golden, soft afternoon light |
| Mood adjectives | lively, relaxed, professional, peaceful, festive, casual, focused |
| Hedging | it looks like, it appears, it seems, I would guess, probably |
| Quantities | a group of, a couple of, several, a handful of, two or three |
Recovering from a Bad Start in Task 3
| Situation | What to Do |
|---|---|
| You blank at start | Use a stall opener: “Okay, so this picture shows…” then begin |
| You repeat “I can see” | Switch to “there is”, “there are”, “in the corner…” |
| You finish at 40 seconds | Add mood and atmosphere: “Overall the scene feels…” |
| You feel rushed at 55 sec | Close with a one-line summary: “It feels like a peaceful afternoon.” |
| You realise you used past tense | Self-correct: “Sorry, the woman IS walking, not WAS walking.” |
Practise on the Real CELPIP Interface
Marvel Education’s CELPIP practice software replicates the real test interface:
- A bank of 200+ realistic Task 3 scene images
- 30-second prep timer matching the official one
- Microphone capture auto-stopping at 60 seconds
- Replay alongside the original image
- AI feedback on present continuous accuracy and vocabulary range
- CLB 10 model responses written by Marvel Education’s CELPIP teachers
- Paired Task 3 and Task 4 practice on the same image for cross-task fluency
Start Free Trial of Marvel CELPIP Practice Software
Connecting Task 3 to Task 4
This is one of the highest-leverage habits in CELPIP Speaking: treat Tasks 3 and 4 as a single 2-minute exercise on the same image. Most students treat them separately and miss this advantage.
In Task 3, you describe what is currently happening. In Task 4, you predict what will happen next. The same image stays on screen for both. That means every detail you notice in Task 3 prep doubles as raw material for Task 4 predictions.
How to Mine Task 3 for Task 4
| Task 3 Observation | Task 4 Prediction You Can Make |
|---|---|
| “The woman is examining apples” | “She will probably buy them and continue shopping.” |
| “The bus is approaching” | “The people waiting will likely board it within a minute.” |
| “A child is laughing with the dog” | “They will probably keep playing for a while until called by the parents.” |
| “The sky looks grey” | “It might start raining, and the market may need to close early.” |
Tip
During Task 3 prep, mentally tag two or three elements as “Task 4 fuel.” Things that imply movement, change, or an unfinished action. A bus pulling in, a child mid-laugh, a shopkeeper unpacking. These become your strongest predictions one minute later.
This cross-task strategy alone has helped many CLB 8 students push into CLB 9 territory because Task 4 stops feeling improvised.
Quick Recap: The Task 3 Playbook
- Spend the 30 seconds counting elements, not writing sentences.
- Map the image into zones: foreground, middle, background.
- Open with a one-sentence big picture line.
- Describe people first (20 sec), then objects and setting (15 sec).
- Use present continuous for actions (“the man IS walking”).
- Use present simple for stable elements (“there ARE tables”).
- Cover 5 to 7 specific elements with detail.
- Use spatial prepositions to organise: foreground, behind, further back, overhead.
- Avoid opinions and predictions. That is Task 4 territory.
- Close with light speculation about mood or context: “It looks like a weekend market.”
Frequently Asked Questions
You get 30 seconds to prepare and 60 seconds to speak. The 30 seconds of prep is spent studying the image, not writing notes. Use the prep window to identify foreground, middle, background, and 5 to 7 specific elements.
Present continuous, almost entirely. ‘The man is walking’, ‘a child is laughing’, ’two people are sitting’. Present simple is fine for stable elements: ‘There are tables’, ‘The sky is blue’. Avoid past tense entirely. Task 3 is about what is happening right now in the image.
No. Task 3 is purely descriptive. You are reporting what you see. No opinions, no judgements, no advice. Save opinions for Task 4 (predictions) and Task 7 (opinions). Mixing functions hurts your Task Fulfilment score.
Aim for 5 to 7 specific elements across foreground, middle ground, and background. You can describe people, objects, weather, colours, positions, and activities. Vague summary descriptions (‘a lot of people doing things’) score lower.
Use hedging language: ‘It looks like…’, ‘It seems to be…’, ‘There appears to be…’. This shows vocabulary range and lets you describe ambiguous elements without making firm claims you might get wrong.
By zone. Start with the big picture (one sentence), then describe foreground (closest), then middle ground, then background. Left-to-right scanning often produces choppy descriptions with no spatial logic. Zones give the listener a clear mental map.
Light speculation is fine and shows vocabulary range. ‘They appear to be a family’, ‘It looks like a weekend market’. Just keep it grounded in visible evidence. Wild speculation about backstories hurts Task Fulfilment.
Yes. Task 4 uses the same image. In Task 4 you predict what will happen next in the scene you just described. So your Task 3 description sets up your Task 4 predictions. Notice details in Task 3 that you can use to predict in Task 4.