TOEFL Listen and Repeat 2026: Shadowing Drills That Push You to Band 6
Listen and Repeat is the first task you face in the 2026 TOEFL Speaking section, and it is the task that catches most students off guard. You get a short sentence tied to an image, a 7-second window to repeat it aloud, and then the recorder auto-advances whether you are ready or not. There are 7 of these in a row, and they set the tone, literally, for the rest of your Speaking score. The good news is that Listen and Repeat rewards a specific skill called shadowing, and shadowing is trainable. This guide explains exactly what raters score, breaks down the four pronunciation pillars that matter, and gives you five drills plus a 2-week plan to lock in band 5 or 6 performance.
1. What the Listen and Repeat task actually is
Listen and Repeat replaces the old TOEFL iBT Independent Speaking task as the opening section of the 2026 Speaking exam. You see an image of a campus or everyday setting (a bookstore, a lab, a library, a dorm common area) and you hear a single sentence spoken by a native English voice. Your job is to repeat that sentence back as accurately as possible within a 7-second recording window.
There are 7 Listen and Repeat items per test, each scored independently on the 1.0 to 6.0 band scale. The average of those 7 scores contributes to the Speaking section band, alongside the 4 Interview task scores. This means each item is worth roughly 1/11th of your overall Speaking band, so a stumble or two is recoverable, but a full breakdown on three or four items will cost you a full band on the section.
The sentences themselves range from 6 to 12 words. They are deliberately conversational: things like "Welcome to the university bookstore," "The library closes early on Fridays," or "Please return the equipment to the front desk." There is no academic vocabulary requirement. The challenge is not comprehension. It is reproduction.
2. The 7-second window and the auto-advance recorder
The recorder starts automatically as soon as the prompt audio finishes. There is no "start recording" button to press. You have 7 seconds from the moment the speaker stops to produce your response. At exactly the 7-second mark the recorder stops, saves the audio, and moves to the next item. You cannot pause, re-record, skip, or go back.
This auto-advance behavior is the single most common source of panic on the task. Students who trained on older TOEFL materials expect a "begin speaking now" prompt and a beep. There is neither. You have to build the muscle memory of starting your response within half a second of the prompt ending.
The takeaway: Treat the silence after the prompt audio as if the recorder is already running, because it is. There is no warm-up breath allowed. Practice starting within 0.5 seconds of the prompt ending.
3. What the 1-6 rubric actually rewards
Raters evaluate each response on a single compressed rubric. Unlike the Interview tasks, there is no credit for content coverage or ideas; the content is fixed. What is being measured is how faithfully you can reproduce a sentence with correct pronunciation, stress, rhythm, and intonation.
| Band | Description |
|---|---|
| 6.0 | Complete, accurate, natural rhythm. Stress and intonation match a native model. No intelligibility issues. |
| 5.0 | Complete and accurate. Minor pronunciation slips. Rhythm mostly natural with one or two hesitations. |
| 4.0 | Complete but with visible stress or vowel errors. Listener has to work slightly to understand. |
| 3.0 | Sentence mostly intelligible but with several errors. Missed words, wrong stress patterns, flat intonation. |
| 1-2 | Unintelligible, significantly truncated, or effectively non-response. |
4. The four pillars: intelligibility, stress, linking, intonation
Band 5 and 6 responses share four properties. Each is trainable in isolation.
Intelligibility
Every word must be clearly produced. This is the floor. If a rater cannot identify a word on first listen, that word is scored as incorrect. Focus on vowel length and final consonants; those are what English listeners actually use to decode words.
Word stress
English is a stress-timed language. Some syllables carry weight, others are compressed. "ReTURN the eQUIPment" lands correctly; "REturn the Equipment" sounds foreign even with perfect vowels.
Linking
Native speakers do not pronounce each word separately. "Welcome to the" becomes "Wel-cum-tuh-thuh" with no gaps. Students who articulate every word separately sound robotic, which drops rhythm scores even when pronunciation is correct.
Sentence intonation
Statements fall on the final stressed syllable. Questions rise. A flat, monotone delivery caps you at band 4 no matter how accurate your words are. Intonation is the cheapest band to earn once you hear it.
5. Five shadowing drills that move the needle
Shadowing is repeating a spoken model while it is still playing, staying a syllable or two behind the speaker. It is the single most effective Listen and Repeat exercise because it forces your mouth to match rhythm and stress in real time. These five drills layer from easiest to hardest.
Play a 7-second audio clip. Repeat the exact sentence immediately after it ends, trying to match tone and rhythm. Do 10 reps per sentence. This is the closest simulation of the actual test and builds start-speed reflex.
Play the audio and speak along with it, starting 1 syllable behind the speaker. Record yourself. Play the recording back and compare your stress peaks to the model's. Do 5 sentences per session.
Before repeating a sentence, write it out and underline the 2 to 3 syllables that carry stress. Read your underlined version out loud with exaggerated stress. Then play the model and shadow. This builds conscious awareness of stress patterns.
Take a sentence and mark the linking spots, the places where a word ends in a consonant and the next starts in a vowel. "Return_the_equipment" has linking at "return-the" and "the-equipment". Rehearse the linked version until it flows without gaps.
Play the audio, wait 3 seconds, then reproduce the sentence from memory. This trains retention for longer sentences and is the closest analogue to the real test condition. Use this drill in the final week before your exam.
6. Sample items with stress marking
Here are three realistic Listen and Repeat sentences with the stressed syllables marked. Read each one out loud three times, emphasizing the marked syllables.
Scene: Campus bookstore
WelCOME to the uNIversity BOOKstore.
Linking: "WEL-cum-tuh-thuh-yoo". Stress on COME, NI, BOOK. Falling intonation on "store."
Scene: Laboratory
Please reTURN the eQUIPment to the FRONT desk.
Linking: "return-thee-quipment". "Please" is unstressed and soft. Sentence falls on "desk".
Scene: Library
The LIbrary CLOses early on FRIdays.
"Early" is weakly stressed. Students often mispronounce "closes" as two equal syllables; the first is the stressed one.
7. The mistakes that cap you at band 3
Starting too late
Half a second of dead air at the start of a 7-second window is 7 percent of your response lost. Students who wait for a beep that never comes often only get 4 or 5 seconds of speech recorded, which truncates the sentence and caps the item at band 3.
Speaking word by word
"Please. Return. The. Equipment." scores worse than a slightly imperfect fluent delivery. Raters penalize lack of natural rhythm heavily. Aim for connected speech even if a single word is slightly off.
Over-correcting mid-sentence
If you mispronounce a word, keep going. Restarting eats your 7-second budget and often means the sentence is truncated at the end. A single pronunciation error is a band 5; a truncated sentence is a band 3 or lower.
Flat monotone delivery
Students who focus so hard on accuracy that they forget to inflect sound robotic. Raters score delivery. Even a correct sentence spoken in monotone caps at band 4.
Shouting or over-projecting
The microphone is close to your mouth. Speaking at a normal indoor volume produces cleaner audio than projecting. Over-projection creates clipping that raters hear as distortion, not confidence.
8. A 2-week practice plan
This schedule assumes 20 to 30 minutes a day. It is enough to move an average test taker from band 3 to band 5 on this task.
| Days | Focus | Drills |
|---|---|---|
| Days 1-3 | Intelligibility and stress | Drill 1 (parrot) + Drill 3 (stress mapping) |
| Days 4-6 | Linking and flow | Drill 2 (simultaneous) + Drill 4 (linking) |
| Days 7-9 | Intonation | All 5 drills, focus on falling/rising patterns |
| Days 10-12 | Full simulation | Drill 5 (delayed) + timed 7-second windows |
| Days 13-14 | Full mock tests | Two full Speaking sections end-to-end |
9. FAQ
What is the Listen and Repeat task on the TOEFL 2026?
Listen and Repeat is the first task type in the 2026 TOEFL Speaking section. You hear a short sentence tied to an image of a campus or everyday scene, then you have a 7-second window to repeat it aloud. There are 7 items per test, each scored on a 1-6 band for intelligibility, pronunciation, stress, and intonation.
How long do I have to repeat each sentence?
You have 7 seconds per sentence. The recorder auto-advances when the window expires, so you cannot pause, re-record, or skip. Most target sentences are 6 to 12 words, which fits comfortably inside 7 seconds at a natural pace.
What does the Listen and Repeat rubric reward?
Raters score intelligibility first, then pronunciation accuracy, word stress, and sentence-level intonation. A band 6 response reproduces the sentence clearly, with correct word stress and natural rhythm. A band 5 has minor slips that do not impede understanding. Band 3 responses are intelligible but contain multiple pronunciation or stress errors.
What is shadowing and why does it help?
Shadowing is the practice of repeating a spoken model out loud while it is still playing, staying a syllable or two behind. It trains your mouth to match native rhythm and stress patterns faster than listen-then-repeat practice. Ten minutes of daily shadowing across two weeks produces a measurable jump in Listen and Repeat scores.
Can I pause the TOEFL Listen and Repeat recorder?
No. The recorder runs automatically during the 7-second window and advances to the next item at the end of that window. There is no pause, no retry, and no way to skip ahead. Test takers should expect the auto-advance behavior and practice with a clock to build the muscle memory.
Listen and Repeat feels intimidating the first time you meet the auto-advance recorder. It rewards a single trainable skill. Two weeks of focused shadowing, stress mapping, and timed 7-second windows will move most test takers from panic to fluent reproduction. Pair this task with the Interview task strategies and you will have the full Speaking section covered.
Practice Listen and Repeat with the real 7-second auto-advance recorder
Our free TOEFLMock speaking tests simulate the exact interface, auto-advance behavior, and scene images used on the 2026 exam.
Start Free Speaking PracticeTest preparation specialist and former classroom instructor. Designs full-length mock content aligned to the 2026 ETS redesign and writes study-plan, format, and score-requirement guides.