Read Aloud
A sentence on screen, twenty seconds, one recording. No preparation and no second take — the shortest speaking task on the test and the one people underrate most.
- 20 sec on the clock, Several per full test
- Feeds your Reading and Speaking subscores
- Built for the current July 2025 DET format, including the new Interactive Speaking task
Read Aloud shows you a written sentence and records you saying it. There is no thinking time, no retake, and the whole thing is over in under half a minute.
Because you are given the words, this is the only speaking task where content is not being judged. That sounds like it makes it easy. What it actually does is strip away everything you could hide behind: there is no clever idea to distract from a hesitant delivery, and no vocabulary choice to show off. All that is left is how you sound saying English.
It is also the speaking task that improves fastest, because the skill it measures — turning written English into fluent spoken English in real time — responds to straightforward daily practice in a way that spontaneous speech does not.
What you actually see
One sentence, a record button, and a very short timer. Some sentences are simple; others are deliberately built with clause structures that are awkward to read cold.
- A single written sentence, typically 10–20 words.
- About twenty seconds — far more than the sentence takes to say, so the timer is not the pressure.
- One recording. There is no preview and no re-record.
- Several appearances across a full test, at difficulty that adapts to how you are doing.
- Sentences that get syntactically harder as you succeed — embedded clauses, longer noun phrases, unusual word order.
The same sentence, twice
The sentence on screen:
"The library keeps its rare maps in a climate-controlled room beneath the main reading hall."
A weak reading goes word by word at an even pace, pauses after maps because that is where the reader ran out of planning, stumbles on climate-controlled and restarts it, and lets the pitch fall at the end of every few words as though each were a sentence.
A strong reading groups the sentence the way an English speaker would: The library keeps its rare maps / in a climate-controlled room / beneath the main reading hall. Three phrases, one falling tone at the very end, stress landing on rare, climate and reading.
The words are identical. What differs is phrasing, stress and a single continuous run — and that is essentially the whole of what this task measures. You do not need an accent you do not have; you need the rhythm.
How Read Aloud is scored
Your recording is analysed automatically against the sentence you were given. Because the target text is known, the system can compare what you said to what you should have said very precisely — which is what makes this task reliable enough for Duolingo to use it repeatedly.
Duolingo has not published the exact acoustic measures involved, so treat any specific formula you read online with suspicion. What is clear from the task design is the shape of it: intelligibility, fluency of delivery, and accurate rendering of the words on screen.
On accent, the important thing to understand is that international English tests are built to accept the full range of educated English accents. What is penalised is unintelligibility, not regional origin. Speaking clearly in your own accent scores better than a strained imitation of one you do not have.
| What the scorer looks at | Why it matters |
|---|---|
| Saying the words that are on screen | Substitutions, omissions and added words all count against the reading. |
| Fluency and continuity | One continuous delivery beats a technically perfect one broken by restarts and long pauses. |
| Intelligibility | Whether a listener can identify each word. This is the pronunciation criterion, not accent-matching. |
| Phrasing and stress | Grouping words into sense units and stressing the right syllables. This is what separates fluent from merely accurate. |
| Recording something at all | A silent or near-silent recording scores nothing. Check your microphone before you begin. |
The mistakes that cost the most
The errors here are mechanical rather than linguistic, which is why the task is so improvable.
Do this
- Scan the whole sentence first and mark, mentally, where the phrase breaks fall.
- Speak at a normal conversational pace. Slowing down to be careful reads as disfluency, not care.
- Keep going if you stumble. A small slip inside a continuous delivery costs less than a restart.
- Sit close enough to the microphone and speak at normal volume — thin, distant audio is analysed as unclear speech.
- Let your pitch fall only at the end of the sentence, not after every phrase.
Not this
- Do not start reading before you have seen how the sentence ends. You will phrase it wrongly.
- Do not restart the sentence. The recording captures the restart and you rarely have time for a clean second run.
- Do not attempt an accent that is not yours. It costs you fluency and buys you nothing.
- Do not read word-by-word-at-equal-stress. English rhythm is uneven, and flattening it is the most common thing that makes a reading sound non-fluent.
- Do not whisper or trail off at the end of the sentence, which is where people run out of breath.
How to actually get better at Read Aloud
This is the single most trainable task on the DET. Fifteen minutes a day for a few weeks visibly changes it, and unlike vocabulary work the gains show up immediately.
Read aloud for ten minutes every day, out of anything
News articles, textbook paragraphs, whatever you already read. The muscle being trained is turning text into speech without planning time.
Record yourself and listen back
This is the step people skip and it is the one that works. You cannot hear your own restarts and flat phrasing while you are producing them.
Mark the phrase breaks before you read
Take a paragraph and draw slashes where the sense units end. Read it respecting them. Then do a fresh paragraph without marking, and see whether the habit transferred.
Shadow a native recording
Play a sentence, then say it immediately after, copying the rhythm rather than the accent. Stress and timing are what transfer.
Practise the awkward structures specifically
Sentences with embedded clauses and long noun phrases are what the adaptive engine serves when you are doing well. Seek them out rather than reading easy prose fluently.
How much this one task moves your score
Read Aloud appears several times in a full test and feeds Speaking, and therefore Conversation and Production.
Its real value is as the cheapest way into your Speaking subscore. The other speaking tasks require you to generate ideas under time pressure; this one hands you the words. If your Speaking number is low and you do not know why, this is where to look first, because it isolates delivery from everything else.
Improvement here also transfers. The phrasing and fluency you build reading aloud show up directly in Listen, Then Speak and the Speaking Sample, where they are worth considerably more.
Read Aloud: questions people ask
Does my accent affect my Read Aloud score?
Intelligibility is what is assessed, not which accent you have. International English tests are built to accept the full range of educated English accents. Attempting an accent that is not yours generally hurts, because it costs you fluency and naturalness while buying nothing.
Can I re-record if I make a mistake?
No. You get one recording per sentence. If you stumble, keep going — a small slip inside a continuous delivery costs far less than an audible restart, and you are unlikely to have time to finish a second attempt cleanly.
How long is each sentence?
Typically ten to twenty words, and they get syntactically harder as you do well. You get about twenty seconds, which is much more than the sentence takes to say — the extra time is for reading it first.
What happens if my microphone fails?
A silent recording scores nothing, so test your microphone before you start the test rather than discovering the problem on the first speaking item. Our practice tests run a microphone check before any speaking task for exactly this reason.
Is Read Aloud easier than the other speaking tasks?
Mechanically, yes — you are given the words, so there is no content to generate. That also means there is nothing to hide behind: delivery is the entire score. Most people find it the easiest to improve and the hardest to fake.
Is Read Aloud still part of the Duolingo English Test?
No. It was removed in July 2025 and folded into the new Interactive Speaking section. This page stays up as a reference, but you will not meet Read Aloud in a live test any more.
Why was Read Aloud removed?
Because reading a supplied sentence aloud measures pronunciation and fluency but not whether you can generate your own language. Duolingo replaced it with tasks built around spontaneous speech, which give a fuller picture of speaking ability.
Does practising Read Aloud still help my speaking score?
Yes, indirectly. The clear phrasing and natural pace it builds transfer straight to the spontaneous speaking tasks that replaced it, so the practice is not wasted — it is simply no longer a scored task in its own right.
How many Read Aloud sentences are there?
Several across a full test, at difficulty that adapts as you go. Sentences get longer and more syntactically awkward — embedded clauses, extended noun phrases — as you perform well.
What happens if I mispronounce one word?
Very little on its own. The assessment covers the whole delivery, so a single unclear word inside otherwise fluent, well-phrased speech costs far less than the restart you would trigger by stopping to correct it.
Should I read slowly to be clearer?
No. Deliberate slowness reads as disfluency rather than care, and it flattens the stress pattern that makes English intelligible. Normal conversational pace with natural phrasing scores better than careful, even-paced reading.
Can I read the sentence before recording starts?
Yes, and you should. The window is far longer than the sentence takes to say, and the extra time exists so you can see where the clause breaks fall before you commit. Starting to read a sentence whose shape you have not seen is how phrasing goes wrong.
Does breathing or hesitating count against me?
Natural breathing at phrase boundaries is exactly what fluent speech sounds like. What costs you is hesitation mid-phrase, audible restarts, and long silences — all of which signal that you are decoding rather than reading.
Do I need a headset, or is a laptop microphone enough?
A built-in microphone is usually fine provided you are close enough and the room is quiet. Thin, distant or echoing audio is analysed as unclear speech, so the room matters more than the hardware.
How quickly can I improve at Read Aloud?
Faster than any other speaking task. Ten minutes of daily reading aloud, recorded and listened back to, visibly changes phrasing and fluency within a few weeks — and those gains transfer straight into the harder speaking tasks where they are worth more.
Where Read Aloud sits in a full test
No task on the DET is scored in isolation. The adaptive engine is building one picture of your English out of every response, so it helps to know how much of the hour this particular task accounts for and which of the others are drawing on the same ability.
The table below is the whole test. Your current page is highlighted.
| Question type | Time | How often | Subscores it feeds |
|---|---|---|---|
| Read & Complete | 3 minutes | 3–6 times | Reading, Writing |
| Read & Select | 5 seconds | 15–18 times | Reading |
| Fill in the Blanks | 20 seconds | 6–9 times | Reading, Writing |
| Listen & Type | 1 minute | 6–9 times | Listening, Writing |
| Speak About the Photo | 90 seconds | at least once | Speaking |
| Interactive Speaking | 35 seconds each | 6–7 questions | Speaking, Listening |
| Write About the Photo | 1 minute | at least twice | Writing |
| Interactive Reading | 7 minutes | 2 sets | Reading |
| Interactive Listening | 6 minutes | 2 sets | Listening, Speaking |
| Speaking Sample | 3 minutes | once | Speaking |
| Writing Sample | 5 minutes | once | Writing |
Two things are worth reading off that table. The first is that the tasks feeding Reading and Speaking are not only this one — improvement transfers, so time spent here shows up elsewhere. The second is that the tasks taking the largest share of the hour are not the ones most people practise most.
Question types that use the same muscles
Listen, Then Speak
You hear a question — it is never shown as text — then speak your answer.
Read the guide → SpeakingSpeak About the Photo
An image appears and you describe it aloud.
Read the guide → SpeakingSpeaking Sample
An extended spoken response to an open prompt. It is sent to institutions alongside your score.
Read the guide →Start a practice test
An hour under real conditions, all eight subscores, and feedback on every written and spoken answer.
Take a test