TOEFL Speaking in 2026 does not include tasks literally called 'Listen and Repeat' or 'Take an Interview' — those are informal labels sometimes used in prep communities to describe the two contrasting demands of the section: Tasks 2–4 ask you to process spoken or written source material and then speak (listen-then-respond), while Task 1 asks you to express personal views in a structured monologue, much like a high-stakes interview answer. Understanding that distinction tells you exactly where your preparation energy should go.
How TOEFL Speaking Is Structured in 2026
The TOEFL iBT Speaking section contains four tasks completed in roughly 17 minutes. ETS scores each task on a 0–4 scale and converts the combined result to a 0–30 scaled score. The tasks fall into two functional categories that map directly onto the 'listen and repeat vs. interview' debate.
| Task | Type | Prep Time | Response Time | Source Material |
|---|---|---|---|---|
| Task 1 (Independent) | Personal opinion / 'interview' style | 15 seconds | 45 seconds | None — your own ideas only |
| Task 2 (Integrated: Campus) | Read + Listen, then speak | 30 seconds | 60 seconds | Reading passage + campus conversation |
| Task 3 (Integrated: Academic) | Read + Listen, then speak | 30 seconds | 60 seconds | Reading passage + lecture excerpt |
| Task 4 (Integrated: Lecture) | Listen, then speak | 20 seconds | 60 seconds | Lecture excerpt only |
Task 4 is the purest form of 'listen and respond': no reading, no personal opinion — you hear a 90-second academic lecture, take notes, and reproduce the main concept plus two supporting examples in 60 seconds. Tasks 2 and 3 blend reading and listening before your spoken response. Task 1 stands alone as the opinion-delivery task where no external material supports you.
What Does TOEFL Speaking Actually Score?
Every task is rated by AI scoring (e-rater) plus at least one human rater on three dimensions: Delivery, Language Use, and Topic Development. The weight of each dimension differs slightly by task type, which is where the listen-and-respond vs. personal-opinion split matters practically.
- Delivery: pace, clarity, and naturalness of speech — penalized heavily for long silences or robotic rhythm on any task
- Language Use: range and accuracy of grammar and vocabulary — complex sentences score higher than simple subject-verb-object patterns
- Topic Development: completeness, coherence, and logical progression — for integrated tasks, this means accurate representation of the source; for Task 1, it means a clear claim with concrete supporting details
A score of 26–30 (the 'Good' band, required by many top graduate programs) demands a rating of at least 3.5 average across all four tasks. Missing the source content in Task 4 or giving a vague, unsupported opinion in Task 1 both pull that average down equally fast.
Which Task Type Is Harder to Score Well On?
Statistically, most test-takers score lower on Task 1 than on the integrated tasks, despite Task 1 having no source material to memorize. The reason is structural: without reading or listening input, there is nothing to fall back on when your mind goes blank under pressure. Integrated tasks give you a content scaffold; Task 1 demands that you generate a complete argument from scratch in 15 seconds of preparation.
Integrated Tasks (2, 3, 4) — 'Listen and Respond'
- Content is provided — you cannot blank on the topic
- Your job is accurate summary + selective emphasis
- Note-taking during listening determines 80% of your response quality
- 60-second response time is generous if notes are solid
- Risk: misrepresenting the source or omitting a key example
Task 1 — 'Take an Interview'
- No source material — all ideas must come from you
- Your job is opinion + two concrete supporting reasons
- 15-second prep must produce a full outline
- 45-second response time feels short without a template
- Risk: vague or overly general answers that score 2 or lower
How to Prepare for Each Task Type Efficiently
Integrated Tasks 2–4: Build a Note-Taking System First
Your spoken response in Tasks 2–4 is only as accurate as your notes. Practice listening to 60–90 second academic passages and writing abbreviated bullet points that capture: the main claim, the first supporting example with one specific detail, and the second supporting example with one specific detail. During your 20–30 second prep time, number those bullets in the order you will speak them. Your response template is then: state the topic (one sentence), explain example one (two sentences), explain example two (two sentences), close with a connector phrase. That structure fills 60 seconds precisely if each sentence runs 8–10 words.
For Task 4 specifically, the professor almost always signals key terms with a definition pause ('this is called...', 'what researchers refer to as...'). Write that term verbatim in your notes — using the professor's exact terminology in your response signals strong topic development to raters.
Task 1: Use a Fixed Opinion Template Every Time
Task 1 prompts follow predictable patterns: preference between two options, agree/disagree with a statement, describe a quality or skill. Build one reusable template and drill it until the structure is automatic. A reliable format: State your position (5–7 words). Give Reason 1 with a personal or hypothetical example (15–20 seconds). Give Reason 2 with a different angle, more specific (15–20 seconds). Close with a single sentence restating your position. This uses your 45 seconds without padding.
Do not try to choose 'the right answer' on Task 1 — there is no correct position. Students who spend their 15-second prep time debating which side to take almost always start speaking with no plan and trail off mid-sentence. Pick the side for which you can generate two concrete examples in under five seconds, then commit fully.
Record yourself answering 10 Task 1 prompts back-to-back, then listen back and flag every filler ('um', 'like', 'you know'). Fillers below 2 per response are acceptable; above 4 per response they actively lower your Delivery score. On Memo Chat+, you can practice timed speaking tasks and get instant feedback on fluency patterns before test day.
Score Targets and Time Allocation for 2026 Prep
If your target scaled score is 24 (the common minimum for graduate admission), you need an average raw score of approximately 3.0 across all four tasks. If you are scoring 3.5 on integrated tasks but 2.0 on Task 1, you cannot reach 24 — the math does not work. Balance your preparation time accordingly.
| Target Scaled Score | Required Average Raw Score | Realistic Strategy |
|---|---|---|
| 20–21 | 2.5 average | Focus on note-taking accuracy for Tasks 3–4; use template for Task 1 |
| 24–25 | 3.0 average | Reduce filler words in Delivery; add specific examples in Task 1 |
| 26–28 | 3.5 average | Complex sentence variety; precise academic vocabulary in all tasks |
| 29–30 | 3.8–4.0 average | Near-native fluency, zero content errors in integrated tasks, sophisticated Task 1 reasoning |
The TOEFL Speaking section rewards systematic preparation over natural talent. Whether you are drilling the listen-and-respond structure of Tasks 2–4 or rehearsing a locked-in template for the opinion task, the mechanism is the same: reduce cognitive load on test day so that limited mental bandwidth goes entirely to content quality and clear delivery — not to figuring out what format to use while the clock runs.