Skip to content
celpip9

Speaking · Task 3 of 8

Describing a Scene

One of 8 tasks, judged as a whole rather than question by question.

How long you get

30 seconds to prepare, then 60 seconds of recording

Not published by the test provider. Check it against the free Starter Set.

What this section does to you

You will speak your answer to each task aloud, and it will be recorded.

Each task gives you preparation time, then recording begins on its own and stops on its own. There is no retake and no way to hear your answer back.

Each task is judged as a whole against the published Performance Standards dimensions, not on a checklist.

What is judged

Each task is judged as a whole on these dimensions, published as part of the Performance Standards. The level-by-level descriptors behind them are not published, so nobody — us included — can tell you exactly where one level ends and the next begins.

  • Content/Coherence
  • Vocabulary
  • Listenability
  • Task Fulfillment

Source: CELPIP Performance Standards for Writing and Speaking (PDF, © 2023 Prometric)

How to approach Task 3

Your job is to describe what is happening in the picture. It helps to imagine you are describing it for someone who has not seen it. With that person in mind, it is clear that a list of objects is not enough. Say who is there, what they are doing, where things are in relation to each other, and which small details bring the scene to life. That imagined person would need to follow you from place to place without getting lost, so the order you choose matters as much as the words you use.

Step by step

  1. Scan the whole picture before you plan

    When preparation time starts, look over the entire scene quickly and ask yourself where this is and what the main activity is. This gives you the big picture for your opening. Every detail you add later then fits inside that frame.

  2. Choose a route through the picture

    In your head, divide the scene into areas: front and back, side to side, or the centre and then the edges. Pick the division that suits this picture and settle the order now. That way you never have to stop partway through to decide where to look next.

  3. Pick the details worth saying in each area

    For each area, find the people and what they are doing. Then choose a few specifics that make them vivid, such as a colour, a facial expression or an object in someone's hands. A clear action with a concrete detail says more than a long run of things you only name.

  4. Open with the scene, then walk the route

    When recording begins, set the scene briefly: the place and the main activity. Then move from area to area. Use words that place things in space and words that signal each move, so it is always clear which part of the picture you are describing. Describe the actions as things happening now, because you are narrating what is in front of you.

  5. Go deeper, then close

    If you finish your route and there is still time, go back to an area and add finer details: something happening in the background, how someone seems to feel, what might happen next. Have a short closing thought ready, such as the overall mood or what the scene seems to be about. Use it when your route runs out or time feels short, so your answer ends on purpose instead of just stopping.

Using the time

Use a brief moment at the start of preparation to take in the whole scene. Spend most of the preparation choosing your route and picking the details for each area. Use the last moment to run your opening through in your head. During recording, keep the opening short and spend most of your time walking through the areas. Save your closing thought for the end. The clock does not stop, so don't pause to re-plan. If you lose your place, move straight to the next area on your route.

What usually goes wrong

  • Naming things instead of describing a scene

    You can slip into naming object after object without saying what anyone is doing. That leaves out what the task asks: what is happening. If you hear yourself naming a thing on its own, tie it right away to a person, an action or a location before you move on.

    In your feedback: Off prompt

  • Jumping around the picture

    Your eye catches something interesting and you leap to it, then jump back, and your description loses its thread. If you notice you have left your route, finish the sentence you are in. Then name the area you are returning to and carry on from there.

    In your feedback: No structure

  • Running dry before the recording ends

    Covering the obvious things quickly can leave you with silence and time still on the clock. When you reach the end of your route early, don't stop. Zoom in on the area with the most going on and describe smaller details, feelings or likely next moments.

    In your feedback: Incomplete

  • Stalling on a word you can't find

    When you don't know the name of an object, searching for it can create a long silence. Describe the object by its shape, its material or what it is used for, and keep going. The object still comes across, and your flow stays intact.

    In your feedback: Long pauses

Checking your own work against what is judged

Content/Coherence
Check whether your answer had a scene-setting opening, a route that was easy to follow and a closing thought. Also check whether each area included specific details rather than bare labels.
Vocabulary
Listen for the same word used again and again, especially the words you use to say something exists or where it is. Check whether you used precise action verbs and exact names for objects, and varied words for location.
Listenability
Notice where fillers or silences showed up. Moving from one area to the next is a likely spot, so check how smooth your moves were. Check also that your pace stayed steady and your key words came out clearly.
Task Fulfillment
Check that you kept to what is happening in the picture, meaning its people, actions and setting, and didn't drift into unrelated stories. Also check that you kept speaking until the recording stopped.

Drafted by an AI writer and passed by a separate AI editor after one rewrite, checked against our content rules: structure and approach only, no wording to reuse. How our content is made

The full lesson in the app adds drills and a self-check →

The rest of Speaking

  1. Task 1Giving Advice
  2. Task 2Talking about a Personal Experience
  3. Task 4Making Predictions
  4. Task 5Comparing and Persuading
  5. Task 6Dealing with a Difficult Situation
  6. Task 7Expressing Opinions
  7. Task 8Describing an Unusual Situation

How the Speaking component works →

Reading about it only gets you so far

You can read what Task 3 asks. What you cannot read is what it feels like to meet the section's own mechanic while the clock runs. The test drive is where you do that, untimed, unscored and free of your quota.