The two tasks
They are officially called Write About the Photo and Speak About the Photo. (People search for "describe the image", which is why you are probably here — but these are the names Duolingo uses.)
| Write About the Photo | Speak About the Photo | |
|---|---|---|
| Preparation | None | 20 seconds |
| Time | 1 minute | 90 seconds |
| How many per test | 3 | 1 |
| Scored on | Content, discourse coherence, grammar, lexis | The same four plus fluency and pronunciation |
| Feeds | Literacy and Production | Conversation and Production |
The photo stays on screen while you write. You get one recording attempt on the spoken one.
Why the same description should not be built the same way
The written task is scored on four criteria; the spoken task on six. The two extra ones — fluency and pronunciation — change what a good answer looks like.
In Write About the Photo you have sixty seconds and a keyboard, and nothing is measuring your pace. Spelling and punctuation are inside the criteria. So the written answer should be compact and correct — a few full, well-punctuated sentences, not fragments.
In Speak About the Photo you have ninety seconds and a clock, and stopping early costs you. Here the enemy is not the wrong word — it is the silence while you search for the right one. So the spoken answer should be longer, and it should keep moving, even when you have to work around a word you do not have.
The opening sentence to delete
Duolingo's own guide singles out one thing candidates do, and tells them not to: long introductions of the type "This picture shows a situation that might be happening in everyday life, and I think it could be about…"
In a one-minute task, that sentence is a quarter of your answer and it contains no information about the photo. Every criterion it touches, it lowers — content, because nothing has been said; coherence, because the ideas have not started; lexis, because those are all general words.
Start with the thing in the picture:
| Instead of | Write |
|---|---|
| This image shows a scene which appears to depict… | An older man is repairing a bicycle on the pavement outside a shop. |
| In this photo, we can see there are some people… | Three children are queuing at an ice cream van. |
The language that scores
The guide's other tip is short: say more than just naming things. Describing in detail is what separates a mid answer from a high one, and there are four kinds of detail worth having ready.
1. Precise nouns and verbs, not general ones
This is the lexis criterion in one line — it asks for lexical sophistication, meaning more precise or less common words, not longer ones.
- a man is doing something to a bike → a mechanic is tightening the chain on a bicycle
- there are some things on the table → a laptop, a half-empty coffee cup and a stack of papers are on the desk
- she is very happy → she is grinning, holding the trophy above her head
2. The present continuous, because the photo is a moment
A woman is waiting at a bus stop is describing the photograph. A woman waits at a bus stop is describing a habit. The tense choice is part of the grammar criterion, and the continuous is almost always the right one here.
3. Position, so the description has a shape
In the foreground / behind her / on the left-hand side / in the background / just outside the frame. These do real work for discourse coherence, which asks whether ideas follow a logical order — and a description with no spatial order does not have one.
4. One inference, marked as an inference
It looks like early morning, because the street is empty and the light is low. They seem to be celebrating something — everyone is dressed formally. That is development, which is explicitly in the content criterion: are the ideas supported with relevant detail?
Hedge it (it looks like, they seem to be, presumably) — that is accurate, and it is more sophisticated grammar than a flat statement.
A structure that fits the clock
Not a template to memorise — a running order, so you never spend your seconds deciding what to say next.
- The main subject, immediately. Who or what, doing what, in one sentence. No windup.
- Where they are, and one detail of the setting.
- What else is in the frame — the background, other people, the objects that matter.
- One inference, hedged: what is probably happening, and why you think so.
- (Spoken only) Keep going until the time runs out — add another detail, or say what might happen next.
For the written task, four solid sentences in this order, correctly punctuated, will comfortably outscore twice as many words of vague commentary.
The same photo, twice
A photo of a busy market stall.
Nothing here is wrong. That is the point — accuracy is not what is missing. There are no precise nouns, no position, no development, and the first sentence says nothing at all.
Same photo, same minute. The difference is not vocabulary you would need to memorise — weighing, counter, crates, shoppers are ordinary words. The difference is that the second answer looks at the photo and reports what is in it, which is the entire task.
Frequently asked questions
How long do I get to describe the photo on the Duolingo English Test?
Write About the Photo gives you one minute to type, with no preparation time, and there are three of them per test. Speak About the Photo gives you 20 seconds to look at the image and 90 seconds to speak, and there is one per test. The photo stays on screen while you answer, and the spoken task allows only one recording attempt.
How should I start my DET photo description?
With the main subject of the photo, in the first sentence. Duolingo's own guide specifically warns against long introductions such as 'This picture shows a situation that might be happening in everyday life...' — in a one-minute task that is a quarter of your answer and contains no information about the image.
What tense should I use to describe a photo?
The present continuous, in most cases: 'A woman is waiting at a bus stop' describes the moment in the photograph, while 'A woman waits at a bus stop' describes a habit. Tense choice is part of the grammar criterion, so the distinction is worth getting right.
What if I don't know the word for something in the photo?
Describe it instead of skipping it — the official tip says exactly this: use easy words to say what the thing looks like or what it does. Skipping an important object costs you on content, and stopping to search for the word costs you on fluency in the spoken task. Working around a missing word protects both.
Are the two photo tasks scored the same way?
No. Write About the Photo is scored on content, discourse coherence, grammar and lexis. Speak About the Photo is scored on those four plus fluency and pronunciation. That is why the spoken answer should keep moving and use the full 90 seconds, while the written one should be compact, complete and correctly punctuated.