TOEFL 2026 guide
The Speaking section, task by task
About eight minutes, two task types, and no preparation time worth the name. What the 2026 Speaking section actually measures is narrower than most students assume — which is good news, if you know what it is.
The shape of the section
Speaking is now the shortest section on the test, at roughly eight minutes. It has two parts: repeating sentences you hear, and answering questions in an interview format. The old independent and integrated essays-out-loud are gone.
| Task | What happens | What is scored |
|---|---|---|
| 1 · Listen and Repeat | You hear a sentence and say it back | Pronunciation, stress, fluency — not ideas |
| 2 · Take an Interview | Short questions, about 45 seconds each | Whether you answer, in natural connected speech |
This section does not test your opinions
There is no correct view, no essay structure to memorise, and no reward for sophistication of content. Students who prepare Speaking by writing model answers about technology and the environment are preparing for a test that no longer exists.
Task 1: what “repeat” actually rewards
Repeating a sentence sounds trivial and is not. It isolates the parts of speech that carry intelligibility: where the stress falls, whether the vowels are the right length, whether the sentence has one shape or five disconnected words.
- Copy the RHYTHM before the sounds. English stresses content words and swallows the rest; a learner who gives every word equal weight sounds wrong even with perfect individual sounds.
- Do not slow down to be careful. Careful and slow is scored worse than fluent and slightly imprecise, because the thing being measured is connected speech.
- Start speaking immediately. Hesitation before the first word costs more than a mispronounced consonant in the middle.
- If you miss a word, keep the shape and keep going. A sentence delivered whole with one word wrong beats a sentence that stops.
Task 2: answer the question that was asked
The interview format gives you short questions and about 45 seconds each. The commonest failure is not weak English — it is answering a slightly different question, or spending thirty seconds on preamble before reaching the answer.
- Answer in the first sentence. Then support it. Never build up to the answer.
- One reason developed beats three reasons listed. Forty-five seconds is not long.
- Use a concrete example. “When I moved to a new school” gives you fluent, personal language; abstractions make you hesitate.
- Finish your sentence when time is nearly up. A cut-off clause reads as a loss of control.
Fluency is trainable; vocabulary range is not, in the time you have
If you have weeks rather than months, spend them on delivery: speaking without long pauses, linking words together, keeping a steady pace. Those move a Speaking score faster than any amount of new vocabulary, because they are what the scale actually describes at the bands most students are moving between.
On accent
You are not scored on having an accent. You are scored on being understood. Those are different, and conflating them makes students self-conscious in a way that directly damages fluency — which IS scored.
The features worth working on are the ones that change meaning or cost comprehension: word stress in the wrong place, final consonants dropped, and vowel length collapsed. A Korean, Japanese or Chinese speaker who fixes those three is intelligible, and intelligible is what the band descriptors ask for.
Notes on this section
Shorter pieces, each about one thing — a single fact about the 2026 format, or a single pattern that costs people marks.
There is no preparation time →Eleven questions, windows of 8 to 45 seconds, and no planning phase anywhere.