TOEFL Speaking Section 2026: Tasks, Timing, and Practice

The 2026 TOEFL Speaking section uses two task types — Listen and Repeat and Take an Interview — completed in one short session. A full practice section combines seven Listen and Repeat sentences with four interview questions, recorded once through a browser microphone with no replay and no rewrite.

In this guide

1. What the 2026 Speaking section includes
2. Task-by-task timing
3. Why one-take recording matters
4. What to practice first
5. How a practice report should be used
6. Try this yourself
7. Frequently asked questions

What the 2026 Speaking section includes

Listen and Repeat asks you to hear a campus sentence once and reproduce it after a beep. The sentences get longer as the set continues, and a small scene image may give context, but the wording is not shown as a script you can read. The skill being measured is immediate spoken recall: content words, word order, and a steady, understandable delivery inside a short window.

Take an Interview asks you to answer four heard questions on one familiar topic. Each answer has a 45-second response window. The questions typically move from a factual prompt, to an experience, to an opinion, to a more supported comparison. You do not get a planning screen, and you should not spend the first ten seconds restating the question.

A full section is eleven items, not four old Independent and Integrated tasks. If a study resource still uses those labels, treat it as outdated for current practice. See TOEFL Speaking 2026 vs the old format before you reuse old templates.

Task	Items	Response window	Main skill
Listen and Repeat	7 sentences	8, 8, 10, 10, 10, 12, 12 seconds	Accurate, intelligible repetition
Take an Interview	4 questions	45 seconds each	Direct, developed spoken answers
Full section	11 items	About 8 minutes of speaking work	Both skills under one session

Task-by-task timing

The recording windows are short on purpose. In Listen and Repeat, an 8-second sentence leaves almost no room to restart. A 12-second sentence is long enough to finish a complex campus instruction, but only if you start immediately after the beep. The useful habit is not speed for its own sake; it is beginning on time and keeping the sentence moving.

In Take an Interview, 45 seconds is long enough for a complete answer and short enough that wandering costs you the ending. A practical split is about 5 seconds to state the answer, 10 seconds for a reason, 20 seconds for one example, and 10 seconds for a result or closing point. If your first word arrives at second 8, the example is already compressed.

Practice should include the transition from one item to the next. Isolated drills are useful, but only a full 11-item run shows whether Listen and Repeat fatigue makes the first interview answer slower or thinner. That fatigue effect is one of the reasons a full Speaking practice session is different from repeating a single sentence in a quiet room.

Why one-take recording matters

Replaying a prompt or polishing a second recording changes the skill being measured. The current section rewards what you can produce after one listen or one heard question. If you pause the timer, write a script, or re-record until the answer sounds smooth, you are practicing editing, not speaking under exam conditions.

Browser microphone practice is closer to test day than uploading a file you prepared elsewhere. You hear the item, wait for the signal, and speak once. After the section, review evidence instead of guessing: which words dropped, how long the first pause was, and whether the second half of an interview answer stayed on the question.

This is also why “practice until it sounds perfect” is a weak default. A slightly imperfect first take that starts on time is more informative than a polished third take. The TOEFL Speaking checker is useful only when the attempt was one-take; otherwise the report describes a different task.

What to practice first

Start with a clear first sentence. In Listen and Repeat, that means preserving names, places, verbs, and numbers even if a small function word is missing. In Take an Interview, that means answering the question in the first sentence instead of warming up with a general comment.

Then isolate one delivery problem at a time. Fluency work should target time to first word, longest pause, or filler count — not all three in the same attempt. Intelligibility work should target content-word stress and finished endings, not a new accent. Organization work should target a direct answer plus one example, not a more impressive vocabulary list.

A good sequence for most students is:

1. One Listen and Repeat set focused on content words.
2. Four interview answers using the same four-part shape.
3. One full section to see which skill drops under time.
4. The next day, only the weakest of those three.

Do not chase a higher words-per-minute number if clarity falls. Fast, unclear speech is harder to follow than a controlled pace with complete phrases. How TOEFL Speaking is scored in 2026 explains why the report separates fluency from intelligibility.

How a practice report should be used

A useful Speaking report does more than return a single band. It should show item-level estimates, the dimensions that applied to each task, and the evidence behind the rating: word alignment for Listen and Repeat, and transcript plus timing metrics for interview answers.

Use the report as a filter. If Repeat Accuracy is the lowest bar, the next session is phrase grouping, not interview storytelling. If Organization is the lowest bar, the next session is a shorter opening plus a concrete example, not pronunciation drills. If time to first word is late on every interview item, prepare reusable openings by question type before you record again.

Treat the 1.0–6.0 figure as orientation only. Two students can land on the same practice band with opposite profiles: one accurate on Listen and Repeat but thin on interview support, another fluent in interview answers but dropping key nouns in repetition. The next study choice comes from the profile, not from the combined number.

TOEFLGRADER practice scores follow public-style criteria and an indicative mapping. They are not official ETS results and cannot guarantee a test-day Speaking score.

Common mistakes that waste a full-section attempt

• Restarting a Listen and Repeat sentence after a small slip, then missing the content words at the end.
• Restating the interview question for eight seconds instead of answering it.
• Saving the example for “if there is time,” then hitting 45 seconds with only a general opinion.
• Reviewing only the overall band and repeating the same full section without a focus rule.
• Using old Independent Speaking templates that assume 15–30 seconds of note-taking.

Each of those mistakes is visible in evidence: a late start, a long first pause, a dropped noun, or an answer that never returns to the question. Fix the visible pattern before you add a new technique.

Try this yourself

Complete one full 11-item Speaking section on the TOEFL Speaking practice page or through the TOEFL Speaking checker. After the report, write one sentence that names the weakest dimension and the next isolated drill. Do not start a second full section until that drill is done.

Frequently asked questions

How long is the 2026 TOEFL Speaking section?

The speaking work itself is about eight minutes across eleven items: seven Listen and Repeat sentences and four 45-second interview answers. Brief transitions sit around those recordings, so a realistic practice block is a short, uninterrupted session rather than a long old-format Speaking test.

Is there preparation time before each question?

No. Listen and Repeat starts after one listen and a beep. Take an Interview starts when the question ends. Practice should train an immediate start, not a written outline.

Should I practice the two tasks separately or together?

Both. Isolated sets build the specific skill; a full section shows whether accuracy or organization holds when you are already tired. Alternate rather than choosing only one mode.

Are practice scores the same as official TOEFL Speaking scores?

No. Practice reports use 0–5 item criteria and an indicative 1.0–6.0 mapping to guide the next drill. They are not ETS scores.

What should I do if Listen and Repeat is much weaker than the interview?

Spend the next two sessions on one-listen repetition and content-word review, then return to a full section. Improving the weaker task moves the combined practice band more efficiently than extra interview storytelling.

Related reading: how TOEFL Speaking is scored in 2026, Listen and Repeat explained, and a 45-second interview strategy.