シャドーイング練習: How does AI work? - 動画で英語スピーキングを学ぶ
レッスンを作成中...
1
Hey, it's Esther.
2
My friend took this video while riding in a car.
3
Watch the steering wheel turn and see if you notice something strange.
4
There's no one driving.
5
This is what's sometimes called a self -driving car.
6
Instead of a person driving, the car is controlled by a computer system with artificial intelligence.
7
You've probably heard about artificial intelligence or AI.
8
Like maybe you've heard that AI can chat with you online.
9
You type a question and it answers.
10
Or maybe you've heard that AI can make pictures, like this AI image of a made -up animal.
11
Someone named Damari is curious about AI.
12
Let's give Damari a call now.
13
Hi, Esther.
14
Hey, Damari.
15
I have a question for you.
16
How does AI work?
17
That's a great question.
18
AI is changing so fast that we can't cover everything in one video.
19
I just mentioned a few things that AI can do today.
20
Drive cars, chat, and make pictures.
21
These things are really different.
22
You might wonder how AI does them all.
23
But if I told you my friend could do all these things, that wouldn't be so surprising.
24
People can learn to do different things, and learning is a big part of how AI works.
25
So let's start with some ways that you learn.
26
Like how this baby is learning.
27
They're making a mess, but watch closely.
28
What do you think this baby is learning?
29
Now would be a good time to pause the video and discuss.
30
Okay, ready?
31
Maybe you guessed that this baby is learning to feed themselves.
32
For any baby, it takes a lot of tries to get food in their mouth.
33
They learn from what works and from what doesn't.
34
Each time a baby tries something, their brain gathers information and starts to notice similarities or patterns.
35
When a baby gets older, they can use what they learned from past experiences, combined with what they're noticing now, to figure out how to eat new foods.
36
Now imagine if, instead of learning like this, you came up with step -by -step instructions for feeding yourself.
37
Like, step one, find food.
38
Step two, pick up food.
39
Step three, put food in mouth.
40
But wait, how do you pick it up?
41
A slice of pizza is not like tiny peas.
42
And picking up soup?
43
Totally different.
44
Just think of other foods.
45
You'd need specific instructions for each one.
46
that might sound silly for a person
47
but step -by -step instructions are one way
48
that computers work people decide what they want a computer to do
49
and come up with step -by -step instructions for the computer
50
to follow you may have heard this called programming ai works differently
51
people still decide what they want the computer to do
52
but instead of step -by -step instructions they come up with programming that helps a computer system learn to do a task.
53
How that works gets really complicated and varies a lot, but one way is a bit like this messy learning.
54
Similar to how a baby gets lots of experiences with food, an AI system is given lots of information.
55
So an AI system designed to make images would get information
56
in the form of pictures photographs drawings all kinds of pictures we're talking thousands and thousands
57
the system is programmed to find patterns in the information like these pictures of cats
58
i see a pattern of having pointy ears
59
and furry tails the ai system can use patterns like these to make new images of cats.
60
But at first, the information is just a digital jumble, and the AI doesn't know what it's looking at.
61
It's a bit like a baby making a mess with food.
62
The program has to try over and over again, making messy mistakes and finding what works better.
63
That's how the AI learns.
64
Let's give the AI system a new task.
65
Make a picture of a unicorn cat.
66
It can use patterns it learns from pictures of cats
67
and patterns it learns from pictures of unicorns to predict what's important to include for something that's part cat and part unicorn.
68
Then it decides how to make a new image.
69
Sometimes AI systems don't get things right.
70
That's why AI systems are programmed to learn from what works and what doesn't in order to improve.
71
Today, some AI systems are able to do complicated tasks like driving,
72
and some can make images of things that seem real but aren't, like this one.
73
Hummingbirds don't actually live inside tulips.
74
This is an image made by AI.
75
This photo was taken in the real world.
76
That's a real hummingbird's nest.
77
See how incredibly tiny it is?
78
This photo and this AI image both seem real.
79
But the AI one is not what happens in real life.
80
That's why it's important to look up other kinds of information to learn more and check facts.
81
So in summary, AI systems do different things, but the basic way that they work is similar.
82
Instead of following step -by -step instructions, AI systems work by learning.
83
You learn from experiences and use what you learn to figure out how to do new tasks
84
and make decisions that's your intelligence
85
and your intelligence is the inspiration for artificial intelligence both you
86
and ai systems learn from what works
87
and from what doesn't work in order to get better at a task ai is changing fast
88
and there's much more to explore along with impressive things ai
89
can do there are also big concerns like how much energy and water is used to run AI.
90
There's lots of debate about when to use AI and what kinds of tasks to use AI for.
91
Since you're growing up in a world with AI, you can be part of figuring out how we use it in the future.
92
It's good to remember that having AI do a task for you isn't the same as trying it yourself.
93
Your brain still needs to learn by doing things that's part of your intelligence
94
so stay curious that's all for this week's question thanks damari
95
for asking it we want to know what you're curious about it's time to cast your vote
96
we picked three questions that we're thinking about answering
97
when this video's done playing click on the one you're most
98
excited to see answered your vote will help us plan for future mini lessons.
99
There are mysteries all around us.
100
Stay curious and see you next week.
📺 同じチャンネル
A21
Where do bugs go in winter?
KidsDaily Conversation
B11
Why do snakes shed their skin?
American AccentKidsTeensHigh School Students
A24
How is butter made?
American AccentKidsTeensDaily Conversation
B111
Why do we celebrate April Fool's Day?
American AccentKidsTeensDaily Conversation
B13
Why is the sky blue?
American AccentTeensDaily ConversationHigh School Students
B11
Do sharks really want to eat people?
American AccentTeensDaily ConversationHigh School Students
B12
How can you tell if a mushroom is poisonous?
American AccentTeensDaily ConversationHigh School Students
C133
How do bees make honey?
American AccentIELTS PrepTOEFL PrepUniversity Students
B14
Are killer whales really dangerous?
American AccentTeensDaily ConversationHigh School Students
B17
Are vampires real?
American AccentTeensDaily ConversationHigh School Students
B1How is glass made?
American AccentHigh School StudentsUniversity Students
✨ おすすめ動画
B11
100 文
AI in Daily Life: Fun Learning with Randy the Robot (AI for Kids) | AI Courses | Education
KidsDaily Conversation
B11
84 文
How Do Airplanes Work? | Educational Video for Kids by Brain Candy TV
KidsDaily Conversation
A21
30 文
What If TECHNOLOGY Didn't Exist? Life Without Technology | Best Learning Videos For Kids | iWonder
KidsDaily Conversation
B11
125 文
Why Chatbots Understand You (And Why They Don't) | NLP Explained
KidsDaily Conversation
B12
48 文
Robot lesson 1
KidsDaily Conversation
B11
335 文
People Who Were Just Built Different
TeensDaily ConversationHigh School Students
B11
5 文
What is DNA and How Does it Work?
Daily ConversationUniversity StudentsWorking Adults
A1805
14 文
Discover How Ai Will Change the World for You
KidsTeensDaily Conversation
A21
78 文
How The Internet Works? | What Is Internet? | Dr Binocs Show | Kids Learning Video | Peekaboo Kidz
KidsTeensDaily Conversation
B21
166 文
How to Create Viral AI Storytelling Shorts
Daily ConversationWorking Adults
B22
20 文
Is the AI Boom Just Another Bubble?
Daily ConversationWorking Adults
B1268 文
Disturbing Robot Thinks It's Human..
TeensDaily ConversationHigh School Students
このレッスンの語彙とスピーキングのポイント
このB1レベルのスピーキングレッスンは、動画「How does AI work?」を教材にしています。 繰り返し出てくる語は次のとおりです：step, learn, picture, baby, task。 この動画には、シャドーイング用の文が100文、単語が1082語あります。 音声の長さは7:45です。 話す速さは1分あたり約140語で安定しており、シャドーイングしやすいペースです。 単語の84%は英語の頻出3,000語に含まれます。残りは練習の前に確認しておきましょう。
この動画の重要語彙
動画の中で特に難しい単語15語を、発音と意味つきで紹介します。
| 単語 | 発音 | 意味 |
|---|---|---|
| instruction 名詞 | /ɪnˈstɹʌkʃən/ | 指示 |
| unicorn 名詞 | /ˈju.nɪ.kɔɹn/ | 麒麟, 一角獣 |
| artificial 形容詞 | /ˌɑː.tɪˈfɪʃ.əl/ | 人工の |
| messy 形容詞 | /ˈmɛsi/ | 乱雑な, きたない |
| hummingbird 名詞 | /ˈhʌmɪŋˌbɜɹd/ | 蜂鳥, 蜂雀 |
| pea 名詞 | /piː/ | 豌豆, エンドウ |
| furry 名詞 | /ˈfɜɹ.i/ | ケモノ, 獣人 |
| similarity 名詞 | /ˌsɪm.ɪˈlæɹ.ɪ.ti/ | 類似, 類似性 |
| steer 動詞 | /stɪɹ/ | 操縦する, 舵を取る |
| slice 名詞 | /slaɪs/ | 切れ, 一片 |
| pause 名詞 | /pɔːz/ | 休止, 中断 |
| nest 名詞 | /nɛst/ | 巣, ネスト |
| predict 動詞 | /pɹɪˈdɪkt/ | 予言する |
| combine 動詞 | /kəmˈbaɪn/ | 合わせる, 結び付ける |
| soup 名詞 | /sup/ | 汁, スープ |
動画に出てくる句動詞
| 単語 | 発音 | 意味 |
|---|---|---|
| pick up 動詞 | 拾う, 拾い上げる | |
| look up 動詞 | /ˌlʊk ˈʌp/ | 見上げる |
この動画の文法
話し手がよく使っている文型を、動画の実際の表現とともに紹介します。
| 文型 | 動画での表現 |
|---|---|
| 受動態 be + 過去分詞 — 誰がするかより、何が起きるかに焦点を当てる | is controlled · is given · is programmed |
| 現在完了形 have/has + 過去分詞 — 過去の出来事が今も関係している | You've probably heard · you've heard |
| 「might / may」による可能性 might/may + 動詞 — 起こりうるが確実ではないこと | might wonder · might sound · may have |
| 現在進行形 am/is/are + -ing — 今まさに、または最近起きていること | is changing · is learning · They're making |
注意したい発音
話し手はdoesn't, you're, you'veなど、短縮形や弱形を21回使っています。聞こえたとおりの短い形で発音しましょう。
- 「sh」と「zh」の音: instruction /ɪnˈstɹʌkʃən/, artificial /ˌɑː.tɪˈfɪʃ.əl/
- 長い単語（アクセントの位置に注意）: artificial /ˌɑː.tɪˈfɪʃ.əl/, similarity /ˌsɪm.ɪˈlæɹ.ɪ.ti/, differently /ˈdɪf.ə.ɹənt.li/
日本語話者が苦手な音：
- /r/ と /l/ の区別 — /l/ は舌先を歯茎につけ、/r/ はどこにもつけない: similarity /ˌsɪm.ɪˈlæɹ.ɪ.ti/, differently /ˈdɪf.ə.ɹənt.li/, explore /ɪkˈsploɹ/
- /f/ — 「フ」ではなく、上の歯と下唇で出す: artificial /ˌɑː.tɪˈfɪʃ.əl/, furry /ˈfɜɹ.i/, photograph /ˈfoʊ.təˌɡɹæf/, differently /ˈdɪf.ə.ɹənt.li/
- 子音の連続 — 間に母音を入れない: instruction /ɪnˈstɹʌkʃən/, unicorn /ˈju.nɪ.kɔɹn/, hummingbird /ˈhʌmɪŋˌbɜɹd/, steer /stɪɹ/, slice /slaɪs/
この動画での練習方法
- まず声を出さずに動画を最後まで聞き、知らない単語をメモします。
- 通常の速度で一文ずつシャドーイングし、話し手のリズムに合うまで繰り返します。
- 自分の声を録音して元の音声と比べます。instruction, unicorn, artificialなどの単語に特に注意しましょう。
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。