跟讀練習: My 17 Minute AI Workflow To Stand Out At Work - 通過影片學習英語口語

選擇練習模式

每種模式都會開啟全屏練習室。

215 句

如果句子過短或過長，請點選 Edit 進行調整。

In 2025, it will take you longer to read something and comprehend it than the amount of time it took to create it.

So if you think finding the signal within all

that noise is only going to get worse with all

that AI generated regurgitation of posts on LinkedIn and all that SEO optimized blog posts that don't actually contain anything good.

So we have to move up the knowledge work value chain and luckily we can do it with AI as well.

So let's come back to the fundamentals of knowledge work and find where we can get leverage.

Knowledge work can be broken down into four steps.

We have our input, the data, the knowledge, the information that we get,

and then we process it to get some sort of output.

right and then from the feedback of the output we then improve our input right

and over time we're going to spiral upwards we're going to

improve the quality of our knowledge work now most people spend

the majority of their time thinking about how can i outsource my thinking process to ai

and so we're using claude we're using chat to pt or say okay you write this for me, you do that for me.

But we realize the results are not really fantastic.

And why is that?

It's because the input that LLMs have are pretty average, right?

We're averaging down to the level of the internet.

It's got some gold in it, but the majority is pretty average.

So what we want to do instead of focusing on process, we want to focus on using AI to improve our inputs.

Even if nothing changes in the way that we process information, our outputs will be better because we won't have this garbage in garbage out problem.

So let's focus our effort on using AI to help us improve our inputs.

And the 8020 we're going to do here is to add academic papers and research into our input.

Now you might think, okay, first of all, boring.

Second of all, this is practical work, right?

This is not some sort of academic research that we're conducting.

We need practical things to happen.

But think about it.

Do you read Harvard Business Review or McKinsey Quarterly or the MIT Sloan Review?

All of these are aimed to take boring, hard, complex academic papers, make it relevant to working professionals

and deliver it in a way

that we can get something actionable out of them we got

to realize now we don't want to be readers of hbr right we read

so slowly but if you just feed the entire archive to chat gpt immediately it will know

so much more than you right

and we don't want to even be the writer for hbr

because people spend their entire life writing a few insights you know you can just leverage those insights

if you don't want to why spend

that time researching instead what you want to be is hbr itself for your team for your organization meaning

that you are going through trusted resources that are complex

that are nuanced take that translate it into the context

that you're in create frameworks that make that theoretical knowledge

and insight into something that's actionable

and be able to share it deliver it in a way

that busy people can understand so we gotta move up the value chain

and look at the things that used to be reserved to academics scientists authors

that had the resources had the time had the brain power

to sit through long -winded papers now you can too

and let's take an example and let me show you the three tools

that i use to make sure i become hbr for my

team for my organization all right let's use an example say one of the biggest things

that managers face which is the problem of team underperform they are bright people

but they seem like they don't want to do anything they're

always complaining about stuff how do i improve team performance okay

so usually if you get average input you might go to chat gpt

and ask it this question

which will give you some average results you might go to

google it might give you some average inputs let's give it a try

so you know how to improve team performance at work common

question right oh gosh it tells you focus on goals focus

on open communication i mean these are just such empty broad recommendations

that is not going to lead to great output

so what i like to do instead is to use the first tool of today elicit

so this helps you find academic papers

on the question that you ask so let's just copy

and paste this

and see what kind of results we get in elicit it's free you just have to sign up and you'll get

something like this you get the summary of the top four papers that's for free

and then you get a list of the papers

that is relevant now what i like to do is just to make sure what i'm reading is

well cited is peer -reviewed

and not some random thing i like to sort by most

cited i take a look for example team design features

and team performance there seems to be a relationship let's take

a look at this one all right what you can do is then click into semantic scholar

which will usually lead you to the actual paper itself

and in this case it is locked so

if you don't have access you know to a university subscription

or something you might not be able to see this

but no problem we'll just copy and paste the paper name and go into google and see

if we can find a pdf version okay here we go

here's the pdf published by sage let's just double check

that seems to be the right one okay so we'll download this

and then we'll use the second tool of today

which is notebook lm by google

so it this tool unlike chat gpt unlike claude is not using ai on the internet's knowledge

but it's only on the sources that you upload

so you have less likely likelihood of hallucination sign up with your google account

and you'll see something like this so you want to create a notebook

and essentially that just houses your sources

and since we downloaded a pdf we are going to just

upload it here now we haven't even read the paper we don't really know what's in there

so what i like to do is to use it as a table of content first, and this is automatically generated by Notebook LN.

All right.

So now we have a table of contents.

Okay.

Groups is important.

There's classification theory.

Looks like we have to look at characteristics, some team size, team level task design, leadership.

100

Okay.

101

Interesting.

102

Now I don't want to read this yet it's a lot of text

103

and really poorly formatted but you know notebook lm is still pretty much experimental

104

so we will give them some time to figure this out but now we have a general idea

105

let's click back on notebook guide

106

and let's look at the suggested questions since we haven't read it

107

but ai has processed it i usually like to use one

108

of the questions just to give me the basic gist

109

so here we go how to team design feature impact overall

110

team performance let's see what it says all right well looks like there are about three components

111

and we saw that in the table content as well right

112

so there's team composition there's task design and then there's leadership

113

and i love this val notebook lm

114

which is you can click into this to find the actual

115

place in the source where it talks about this idea

116

so even though this giant paper is probably you know tens

117

of thousands of words what i can do is figure out

118

which parts are most relevant to me and actually find the citation of it

119

so i can better understand what is it that they say

120

and let's say you know i find i can't change the

121

team composition i'm not going to fire anyone i'm not going to hire anyone new

122

but i want to change task design and leadership i can focus on these cited areas first.

123

Okay.

124

So that's the first cool part about the Elicit to Notebook LM workflow, but this is only one paper, right?

125

So what I would do again is go back to Elicit and now see if there's any other papers that are interesting.

126

Well, leaders can learn about teamwork developing high -performance teams.

127

That seems interesting.

128

Let's take a look, see if we can find the paper oh here is pdf fantastic

129

so now we don't have to oh no

130

okay bad request that's all right it happens let's go back and let's see a via the publisher

131

oh it's a wiley paper let's see

132

if we can get the pdf all right we can't

133

if you're at work i mean surely your work will help you subsidize getting this paper

134

but even if they don't let's see if we can find it somewhere else

135

let's see if this is the right one fantastic all right here's the second pdf we got

136

We will head back to notebook .lm.

137

and this time we're going to press on the plus sign

138

and again upload it here once that's done processing you'll see down here

139

that now you can chat with both sources but before we do

140

that actually i want to uncheck the previous one

141

because i just want to understand what this new paper is about

142

so let me do a table of content for this one source

143

and i will see that it's generating a new note let's wait for

144

that all right here we go there are five essentials to building high performance teams leadership capable

145

and committed members norms a team structure that has clear mission responsibility goal

146

and framework to get results see this is already

147

so much more clear and specific compared to the Google search results

148

which just got diluted over time of everyone maybe one person starting from the source

149

but eventually secondary sources just dilute the message okay a way to manage team processes

150

and that's the fifth one okay great now let's select both and ask it to what are the key character components,

151

key changes a leader can make in a team to improve performance.

152

Now it will look through both sources and give us an answer.

153

All right. So, okay.

154

Focus on member selection.

155

Let's say that's not relevant to us.

156

Promote task meaningfulness.

157

And this was mentioned in the first source, right?

158

Empower the team with autonomy.

159

I don't know if we've seen this.

160

Okay.

161

So this was in the first document.

162

Great.

163

And it gives us the exact things that we should be doing.

164

Cultivate intra -team coordination.

165

All right.

166

Provide effective leadership, which means guiding

167

motivating

168

and supporting the team let's see details now let's say out

169

of all of this i think for my team the thing

170

that makes sense the most is to promote task meaningfulness i want to empower the team

171

and i want to facilitate team development and of course provide effective leadership so

172

give me a detailed plan with three changes each

173

so we can improve team performance now you have a high level plan of how to do this

174

but it still feels somewhat vague even though we know

175

that the input is good

176

and we're focusing on what academics have found out now we want to do the next part

177

which is we're going to copy this and i'm going to go into claude and say help me create a detailed

178

plan to improve team performance using

179

and i'm going to copy in what notebook lm has shared

180

with me now two seconds we got a 12 month three phase like four module plan to improve team performance

181

So just think about if you were HR or you were a team lead and you share this plan with management, they will be blown out of their socks,

182

but that's not all we need.

183

Let's say out of all of this still sounds like really a lot of complications.

184

So let's ask it, what are the 80,

185

20 changes I can make to my team to drive performance

186

that is dependent on leaders and not the team members to make a change and refer to,

187

and I'm going to ask it to refer to these two studies, which is called a meta -analytic review

188

And now, Claude has a better idea of where I'm getting these things from.

189

And so based on details that Ian knows about these two sources, it gave me the core leader action plan okay

190

so task design optimization seems to be one of them authority distribution second highest impact

191

and contextual support so now from 12 months we're down to three months And it gives us some expected results.

192

If that's true or not, we can wait to see, but it helps us list out things like key successful factors, measurement frameworks.

193

So let's say, help me create a matrix to better design tasks with clear goals

194

and interdependence structures.

195

Here we go.

196

Now we've got a framework that we can use to assess goals, clarity level, we can use interdependence level, we can see where they intersect.

197

And we can even show the team, hey, this is what I'm thinking about for changing how we think about tasks.

198

What do you think and get feedback on it, right?

199

Now, if we simply ask Claude, help me with what I need to improve my team's performance.

200

let's see what we get.

201

Alright, it's asking me a few questions that's actually already improvement since previous GPTS.

202

So let's say work team with teams underperforming and unmotivated.

203

Alright, and see, the plan it gives me is so much more vague, right?

204

Communicate and feedback like I know this right having clear goals I know this I know all of this

205

but it doesn't help help me help me create a framework to set goals let's say

206

and yes smart is good monitoring is good but

207

but as you can see

208

because the input quality is not there we're stuck with these buzzwords that we hear all the time.

209

I am excited and optimistic about the changes coming to knowledge work.

210

We can critically think more.

211

We can actually be more creative and associative and generalist.

212

I think we'll have a great time in this era of AI.

213

Let me know how you feel about it.

214

Check out this video on how I use AI to create McKinsey style visuals and I'll see you in the next video.

215

Bye!

下載應用

流利口語所需的一切

AI 發音評分逐句打分

音標練習掌握每個音

詞彙學習建立你的詞庫

詞彙遊戲邊玩邊學

掃碼

Download on theApp Store Get it onGoogle Play

同頻道

How to Think Fast Before You Speak: Framework Thinking

Business EnglishWorking Adults

FOR YOU

本課的詞彙與口語要點

這節 B2 級別的口語課以影片“My 17 Minute AI Workflow To Stand Out At Work”為素材。影片中反覆出現的詞有：paper, performance, improve, input, source。這段影片共有 215 個句子、2511 個單詞可供跟讀。講話部分時長為 17:29。說話人語速平穩，每分鐘約 144 個詞，很適合跟讀。 88% 的單詞屬於英語最常用的 3,000 詞，詞彙容易理解。

影片中的重點詞彙

影片中最難的 15 個單詞，附發音和釋義：

單詞	發音	釋義
notebook 名詞	/ˈnoʊtˌbʊk/	筆記本 /筆記本, 本子
elicit 動詞	/ɪˈlɪsɪt/	引出
upload 動詞	/ˈʌpˌloʊd/	上載 /上載, 上傳 /上傳
insight 名詞	/ˈɪnsaɪ̯t/	見地 /見地
actionable 形容詞	/ˈæk.ʃə.nə.bəl/	可行動的 /可行動的
dilute 動詞	/daɪˈljuːt/	稀釋 /稀釋
underperform 動詞		表現不佳 /表現不佳
empower 動詞	/ɪmˈpaʊ.ə/	授權 /授權
vague 形容詞	/veɪɡ/	模糊, 糢糊
leverage 名詞	/ˈlɛv.(ə.)ɹɪd͡ʒ/	槓桿 /槓桿
pasta 名詞	/ˈpɑ.stə/	мянфан, 義大利麵 /義大利麵
bore 動詞	/boɹ/	煩擾 /煩擾
garbage 名詞	/ˈɡɑɹ.bɪd͡ʒ/	垃圾, 廢物 /廢物
cultivate 動詞	/ˈkʌltɪveɪt/	耕種 /耕種, 種植 /種植
generalist 名詞	/ˈd͡ʒɛn.(ə.)ɹə.lɪst/	通才

影片中出現的短語動詞

單詞	釋義
check out 動詞	退房
figure out 動詞	弄清楚
find out 動詞	找出, 查明
go back 動詞	回去, 返回
go through 動詞	通過 /通過
sign up 動詞	報名 /報名, 註冊 /註冊

值得反覆跟讀的句子

影片中簡短而完整的句子，可以直接用在日常對話裡：

We're averaging down to the level of the internet.
Let's say that's not relevant to us.
I don't know if we've seen this.

影片中的語法

說話人最常用的結構，並附上影片中的原話：

結構	影片中的用法
被動語態 be + 過去分詞 — 強調發生了什麼，而不是誰做的	can be broken · are aimed · are nuanced
現在完成時 have/has + 過去分詞 — 過去發生但與現在仍有關聯的事	haven't even read · haven't read · has processed
定語從句 who / which + 從句 — 補充說明人或事物	face which is · leadership, which means · studies, which is

需要注意的發音

說話人用了 51 次縮略和弱讀形式，例如 we're, don't, I'm。請按聽到的簡短形式來說。

“sh” 和 “zh” 音: actionable /ˈæk.ʃə.nə.bəl/, associative /əˈsoʊ.ʃi.ə.tɪv/, complication /ˌkɑm.pləˈkeɪ.ʃən/, hallucination /həˌluːsɪˈneɪʃən/, optimization /ˌɑptəmaɪˈzeɪʃən/
長單詞——注意重音位置: actionable /ˈæk.ʃə.nə.bəl/, analytic /ˌænəˈlɪtɪk/, associative /əˈsoʊ.ʃi.ə.tɪv/, complication /ˌkɑm.pləˈkeɪ.ʃən/, generalist /ˈd͡ʒɛn.(ə.)ɹə.lɪst/

中文母語者容易讀錯的音：

/v/ — 不要讀成 /w/，上齒輕觸下唇: vague /veɪɡ/, leverage /ˈlɛv.(ə.)ɹɪd͡ʒ/, associative /əˈsoʊ.ʃi.ə.tɪv/, cultivate /ˈkʌltɪveɪt/, motivate /ˈmoʊ.tɪ.veɪt/
詞尾子音 — 要讀清楚，後面不要加母音: notebook /ˈnoʊtˌbʊk/, elicit /ɪˈlɪsɪt/, cite /saɪt/, upload /ˈʌpˌloʊd/, insight /ˈɪnsaɪ̯t/
/r/ 和 /l/ 的區別: leverage /ˈlɛv.(ə.)ɹɪd͡ʒ/, generalist /ˈd͡ʒɛn.(ə.)ɹə.lɪst/, spiral /ˈspaɪɹəl/, critically /ˈkɹɪt.ɪ.kə.li/, clarity /ˈklæɹ.ɪ.ti/

如何用這段影片練習

先完整聽一遍影片，不要開口，記下不認識的單詞。
用正常速度逐句跟讀，每句重複到你的節奏與說話人一致為止。
錄下自己的聲音並與原聲對比，特別注意 notebook, elicit, upload 這類單詞。

什麼是跟讀法？

跟讀法 (Shadowing) 是一種有科學依據的語言學習技巧，最初開發用於專業口譯員的培訓，並由多語言者Alexander Arguelles博士普及。這個方法簡單而強大：您在聽英語母語原聲的同時立即大聲重複——就像是一個延遲1-2秒緊跟說話者的影子。與被動聽力或語法練習不同，跟讀法強迫您的大腦和口腔肌肉同時處理並模模擬實的講話模式。研究表明它能顯著提高發音準確性，語調，節奏，連讀，聽力理解和口語流利度——使其成為雅思口語備考和真實英語交流最有效的方法之一。

影子跟讀法: 閱讀完整分步指南 →

跟讀練習: My 17 Minute AI Workflow To Stand Out At Work - 通過影片學習英語口語

選擇練習模式

下載應用

同頻道

How to Think Fast Before You Speak: Framework Thinking

推薦影片

Give Me 18 Minutes and I’ll Make you Dangerously Smart (with AI)

How AI is Changing Data Scientist's Workflow

AI-First Playbook: Do a Team's Work With AI (2026) | Peter Yang

Top 1% Career Strategy: How to Get Hired in the AI Era

24 Hours Inside a YC AI Startup

My Full Claude Cowork Setup (steal my workflows!)

What Should Marketers Do If Social Media Disappears Tomorrow? l Keynote In 5

The 7 Skills You Need to Build AI Agents

The importance of continuous upskilling in the workplace

I was wrong about AI replacing VAs. Here's what I actually think in 2026.

How demand for workforce skills will change with automation: A minute with MGI

P1 机器学习的应用【2024公认最好的 | 吴恩达机器学习 | 教程 | Machine Learning Specialization(超爽中英!)】

本課的詞彙與口語要點

影片中的重點詞彙

影片中出現的短語動詞

值得反覆跟讀的句子

影片中的語法

需要注意的發音

如何用這段影片練習

什麼是跟讀法？