Говорение: Разбор практических заданий на уроке
TOEFL iBT Тема 5 13:38 Английское озвучивание · Английский + китайские субтитры (встроенные)
Главы
Транскрипт
The Speaking section gives you about eight minutes and about eleven items, in two kinds of task: listen and repeat, and take an interview. You speak into a microphone at a computer. There is no examiner in the room, and no one reacts to what you say, which is exactly what makes it feel strange the first time. The old independent and integrated speaking tasks, where you prepared for fifteen seconds and spoke for forty-five, are gone from the twenty twenty-six test. Eight minutes is short, so almost all of your preparation is about starting quickly, not about having more to say.
In listen and repeat, you hear a sentence and say it back. This belongs to Speaking, and it is scored on what you say; it is not a listening warm-up. You are being heard on pronunciation, on stress, and on whether you reproduce the whole sentence, not on inventing anything new. The skill is holding a whole sentence in your head long enough to say it, and the training is simple: take any recording, play one sentence, say it back, and lengthen the sentences as you improve. Say it at a normal speed, because slowing down to be careful breaks the rhythm of the sentence, and rhythm is part of what is being judged. And if you miss a word, keep going and finish: a complete sentence with one wrong word is worth much more than half a perfect one.
In the interview task you answer a series of spoken questions, as you would in a conversation. The shape that fits almost every question is answer first, then one reason, then one example. Answer the question that was asked, first. Add one reason. Give one example. Then stop, because one explained point is stronger than three unexplained items. Because the bus is cheaper and I can read on it is one good answer; a list of three items is a weaker one. The shape matters because of the clock: it starts you speaking within a second, and a fast start is worth more than extra content in an eight-minute section.
Three behaviours decide how this task goes. First, a silence is worse than an imperfect sentence. Fillers that real speakers use, well, I suppose, that's an interesting one, buy you a moment and sound natural; long pauses do not. Second, do not recite. A memorised answer is audible: the rhythm changes, the vocabulary jumps a level, and it usually does not quite fit the question, so markers discount it. Third, if you lose your way mid-sentence, repair it out loud, sorry, what I mean is, and carry on. That is what fluent speakers do, and it is marked as fluency, not as failure.
Pause and choose. The opening that serves you is the direct answer with a natural hedge: morning, I think, followed by one reason and one example. The both-sides opening postpones the answer past the point where the clock is friendly. The list abandons the question for inventory. And silence while composing the perfect sentence is the one behaviour the marker can only mark down: an imperfect sentence spoken now beats a perfect one that never starts.
Four things are being judged. Fluency: you keep going, at a natural speed, without long silences. Pronunciation: you can be understood without effort, and that does not mean you sound American. Vocabulary: the right word for the idea, used naturally. Grammar: accurate, with enough variety to sound like speech rather than a list. The standard is being understood, not sounding native. An accent is not an error; sounds that change a word's meaning are.
Word stress matters more than individual sounds. PHOtograph, phoTOgrapher, photoGRAPHic: the stressed syllable moves through the family, the weak vowels change with it, and a listener tracks the stress first, so a misplaced stress can hide an otherwise correct word. Sentence stress carries meaning too: I didn't say she stole it means six different things depending on which word you lean on. And practise the sounds Chinese does not use: final consonants like cost, asked, fifths, the th sounds in think and this, and the l, n and r distinctions. A few minutes a day for a month is enough to change how easily you are understood.
Record yourself and listen back. It is uncomfortable, and it is the single most useful thing you can do, because you will hear pauses, repetitions and unfinished sentences you did not notice while speaking. Practise with a timer running, answering aloud into a microphone; the format is the hardest part for most candidates, and it is free to rehearse. Talk to yourself in English during ordinary tasks: describe what you are doing, argue with a podcast, explain your day. Fluency is built by speaking hours, not by reading about speaking. And remember the trap: a typed answer always flatters you. When you practise writing what you would say, you get no hesitation, no silence to fill, and as much time as you like. Speak it aloud instead.
Start within one second of the prompt ending. The clock is short, and a slow start costs more than a poor sentence. Finish your sentences: an abandoned sentence reads as a breakdown, and a simple completed one does not. Warm your voice up before the test by reading aloud for five minutes; a cold first answer is a wasted one, and in an eight-minute section you cannot afford it. And do not try to use difficult vocabulary. Use the right word. Markers reward precision and hear straight through decoration.
A candidate repeats the library opens at nine perfectly, then freezes on a longer sentence with a reason clause. One sentence was held in the head long enough; the other was not. This supplement adds original worked practice to the preserved notes, covering listen and repeat and take an interview. All the sentences, questions and model answers are newly written teaching examples, not official E T S items or rated responses. Reading them silently cannot certify pronunciation, rhythm or spoken delivery, so say every line aloud, ideally into a recording, before reading its coaching notes. The old speaking tasks are gone from the twenty twenty-six test, so this material follows the two current task families only, and the classroom timings are practice choices, not official limits.
Set A gives short sentences for exact rhythm. The lab closes at five. My cousin studies engineering. These seats are reserved today. I printed the form yesterday. The north entrance is closed. Say each at normal speed. Slowing down to be careful breaks the rhythm of the sentence, and rhythm is part of what is being judged, so a careful slow-motion version of the lab closes at five is worse than a natural one with a soft final consonant. Practise with a partner or a recording, never with your eyes only: cover the page, hear the sentence once, speak, then check.
Set B carries structure. The workshop moved to Room twenty-one because the projector is being repaired: a reason clause. You may bring a notebook, but a printed draft is required: a contrast. The second bus was cheaper and less crowded than the first: a comparison. Hold the whole shape, not the first four words plus a guess. Set C is longer still, and long sentences are spoken in chunks, not word by word: the study-support desk, keeps morning appointments in Room four, while afternoon appointments, move to the library annex. Practise hearing the chunk boundaries, then repeat the chunks in order, which is the same skill the interview needs for long answers. If you miss a word, keep going and finish the sentence, and if a sentence collapses, say it once more and move on.
Where do you usually study, at home or somewhere else? The strong version answers first: somewhere else, usually the library. Then one reason: at home I start checking my phone, and in the library everyone around me is working, so I keep working too. Then one example: last week I finished a full lab report there in one evening. The weak version is a list of unexplained places, and lists of three or four items are weaker than one explained choice. Notice also what the strong version is not: it is not memorised. These worked responses are shapes to reuse, not scripts to recite.
The preference question shows the fillers working: morning, I think, I suppose my head is clearest before lunch, and there are fewer messages arriving; for instance, I write my essays between eight and ten, and that is when they come out best. I think and I suppose buy a moment and sound like speech, and they fit an opinion question exactly; they are not weakness. The advice question starts instantly: practise a little every day rather than a lot once a week. Well, because small daily sessions stay in your memory, and a long Sunday session mostly evaporates. When I learned to type, fifteen minutes a day worked better than a two-hour class. Notice the well doing honest work, and remember that if you lose your way, sorry, what I mean is, repairs the sentence out loud and carries on.
Pause and choose. The first failure is the silence, six seconds of nothing in an eight-minute section, and the second is the recitation, because a memorised answer is audible and usually does not quite fit the question. The paragraph's internal quality never got a chance to matter. The trained alternative starts within one second, answers the question that was asked, and accepts being imperfect: an imperfect spoken-now sentence beats a perfect one that arrives late and off-topic.
The pronunciation families worth deliberate practice are the ones that change meaning. Final consonants: say cost and cos, asked and ask, fifths and fits, worked and work, finishing the last sound, and remembering that in asked the e d is a single t sound after the k, not the syllable ed. The th sounds: think, with air and no voice, against sink; this, voiced and light, against dis. The l, n and r family: light, night, right, where the tongue tip touches for l and n and does not touch for r; then red lorry, yellow lorry, slowly, then at speed. And word stress travelling through a family: PHOtograph, phoTOgrapher, photoGRAPHic, with the weak vowels shifting as the stress moves. A listener tracks the stress first, so a misplaced stress hides an otherwise correct word. In the interview, lean on the word that carries your answer: MORNING, because my head is clearest before lunch.
The routine is small on purpose. Record yourself answering one interview question per day, and listen back; it is uncomfortable, and it is the only honest mirror of your timing and rhythm. Run a repeat-sentence drill: five sentences from any recording, growing length, using news podcasts and repeating aloud in the pause after each sentence. Talk to yourself in English during ordinary tasks: describe what you are cooking, argue with a podcast, explain your plan for tomorrow out loud. And practise with a timer running, because the format is the hardest part for most candidates and rehearsing against a clock is free. A typed answer always flatters you: no hesitation, no silence to fill, unlimited time. Speak it aloud instead, or the practice measures your writing, not your speaking.
Pause and check both. First, no: slowing down breaks the sentence's rhythm, and rhythm is part of what is scored; say it at normal speed and finish the final consonant. Second, the error is the sound that changes a word's meaning: cos for cost lands as a different word, while an accent is not an error at all. The supplement's remaining checks: the answer-reason-example shape buys you a one-second start and one explained point; a silence is worse than an imperfect sentence because it gives the marker nothing to score; the stress moves to the first, second and third syllables across the photograph family, and listeners track stress first; and listening back to a recording shows you the pauses and unfinished sentences you cannot hear while producing speech.