쉐도잉 연습: Episode 72 : Neuralk - AI for Structured Data - 영상으로 영어 말하기 배우기

연습 모드 선택

각 모드는 전체 화면 연습실로 열립니다.

161 문장
문장이 너무 짧거나 길면 Edit를 눌러 조정하세요.
1
Hey, everyone, and welcome back.
2
Today, we're going deep on a company that's tackling, well, a challenge you might not have even realized existed.
3
Yeah.
4
AI for the stuff that actually runs most businesses, structured data.
5
It's like, you know, everyone's obsessed with AI writing poems or making art, but this company is focused on the unsung hero of the data world.
6
Yeah.
7
The spreadsheets and databases we all rely on.
8
OK, so tell me more.
9
Yeah.
10
What exactly does this company do?
11
They're called Neuralc AI, and they're based in Paris, France.
12
Ooh la la.
13
Founded by Alexandre Pescu and Antoine Musnot.
14
Okay.
15
Their whole mission is to build AI specifically designed to understand and analyze structured data.
16
Okay.
17
Particularly in the commerce sector.
18
Paris, France.
19
That's kind of cool.
20
Yeah.
21
It adds a little mystique to the AI world, right?
22
It does.
23
But what's the problem they're solving?
24
Right.
25
Is analyzing spreadsheets really that hard?
26
It's not just about analyzing.
27
Okay.
28
It's about understanding.
29
Got it.
30
Most businesses are drowning in data, but they're only scratching the surface of what they can learn from it.
31
Yeah.
32
Traditional analytics can tell you what happened, but Neuroc .ai is building AI that can predict what will happen and help businesses make smarter decisions.
33
So it's like having a crystal ball for your business data.
34
Kind of, yeah.
35
Give me an example.
36
Okay.
37
What kind of problems can they solve with this AI?
38
Think about a massive online retailer with millions of products.
39
Just keeping their product catalog clean and consistent is a huge task.
40
I can imagine.
41
Imagine trying to make sure every product description is accurate, categorized correctly, and all the images are high quality.
42
It's a logistical nightmare.
43
Oh, I bet.
44
And humans probably make mistakes or things slip through the cracks.
45
Exactly.
46
That's where Neuralg AI comes in.
47
They've built models that can automatically clean up product catalogs,
48
standardize information, and even enrich existing data with additional attributes.
49
Wow.
50
They actually did this for a major European retailer called Akan Retail, and the results were impressive.
51
So instead of having a team of people manually combing through millions of product entries, they have AI doing the heavy lifting.
52
That's got to save them a ton of time and money.
53
Absolutely.
54
And it reduces errors, too.
55
And it reduces errors, too.
56
But it goes beyond just cleaning up data, right?
57
Right.
58
They're also using AI to personalize customer experiences.
59
Exactly.
60
Ooh, that's where it gets really interesting for the shoppers like me.
61
Yeah.
62
Tell me more about that.
63
By analyzing things like purchase history and browsing behavior, all structured data, by the way,
64
their AI can predict what products a customer is most likely to want and recommend them accordingly.
65
So it's like those, you might also like the gestures you see online, but powered by really sophisticated AI that understands your individual preferences.
66
Exactly.
67
And it's not just based on simple keyword matching.
68
Okay.
69
They have a model called a tabular foundation model that actually understands the underlying meaning of product attributes.
70
It's kind of like how some AI can understand images or text.
71
Right.
72
But this one is built for tabular data.
73
Okay.
74
The kind you find in spreadsheets and databases.
75
Wait, hold on.
76
Tabular foundation model.
77
That sounds pretty technical.
78
It is a little bit, yeah.
79
Do you break that down for me and the listeners?
80
Sure.
81
What makes it so special?
82
Think of it this way.
83
Okay.
84
Imagine a chef who can not only read a recipe that's like basic analytics.
85
Okay.
86
But also understand flavor profiles and predict how different ingredients will work together.
87
Yeah.
88
That's what this model can do with data.
89
I see.
90
It can see patterns and relationships that traditional methods would miss.
91
Okay, that's a great analogy.
92
Thanks.
93
So it's not just about crunching numbers.
94
Right.
95
It's about understanding the deeper meaning behind those numbers and using that knowledge to make predictions.
96
Exactly.
97
Okay, I'm starting to see how this could be a game changer for companies.
98
Yeah.
99
What other cool stuff have they done with this tabular foundation model?
100
Another example is in similarity searches.
101
Okay.
102
Imagine you're shopping online for a new pair of shoes.
103
Instead of just relying on basic keywords, this model can understand the underlying meaning of product attributes,
104
like style, color, material, and even brand reputation.
105
So it could suggest shoes that are similar in style, even if they don't share the exact same keywords in the description.
106
Yes.
107
It's about understanding the essence of what you're looking for, not just matching words.
108
Right.
109
This makes product discovery much more intuitive and helps customers find what they really want, even if they don't know the exact terms to use.
110
This is all very cool stuff.
111
Yeah.
112
And I can see how this kind of AI could be applied to so many different industries beyond just retail.
113
You're absolutely right.
114
Yeah.
115
Structured data is.. everywhere.
116
And the potential applications for this technology are vast.
117
Well, listeners, it sounds like Neurok AI is definitely a company to watch for them.
118
Where are they headed next?
119
Well, they recently secured $4 million in funding to further develop their technology, specifically their tabular foundation models.
120
That's a pretty significant investment.
121
What are they planning to do with that funding?
122
It seems they're staying laser focused on the commerce sector makes sense
123
which makes sense given their early success there yeah
124
they're also gearing up to launch the first official version of
125
their model oh cool very soon with a full benchmark release expected later this year oh a benchmark release
126
that sounds interesting what does
127
that mean for people like me who might not be you
128
know super technical think of it like a public test they'll be releasing their model
129
and data to researchers and developers so they can see how it performs compared to other AI models out there.
130
It's a way to prove the effectiveness of their technology and show the world what it can do.
131
So it's kind of like a competition for AI models.
132
In a way, yeah.
133
It's a chance for Neuralc AI to demonstrate how their approach to structured data AI stacks up against the competition.
134
This is all very exciting.
135
It sounds like Neuralc AI is really pushing the boundaries of what's possible with AI and data.
136
Absolutely.
137
They're at the forefront of a very important trend.
138
Yeah.
139
We're moving beyond the era of just collecting data and analyzing it in a basic way.
140
Now it's about using AI to truly understand the meaning behind that data and unlock its full potential.
141
It makes you wonder what the world will look like when AI can truly understand all types of data.
142
Right.
143
Not just structured data.
144
It's a fascinating question to consider.
145
Yeah.
146
Imagine a world where AI can seamlessly integrate information from spreadsheets, databases, text documents, images, even audio and video.
147
Wow.
148
The possibilities are almost endless.
149
That's pretty mind -blowing.
150
But for now, it sounds like Neuralc AI is making incredible strides in helping businesses make sense of the data they already have.
151
Exactly.
152
And their upcoming model launch and benchmark release will be key milestones to watch.
153
Right.
154
I'm very curious to see what kind of impact they'll have on the industry.
155
Me too.
156
I'm definitely adding Neuralc AI to my list of companies to keep an eye on.
157
Well, listeners, that brings us to the end of this deep dive into Neuralc AI.
158
But don't worry, the learning doesn't stop here.
159
We'll be back tomorrow evening with another exciting deep dive into a new topic.
160
Make sure you subscribe and hit that notification bell so you don't miss it.
161
Until then, keep exploring, keep learning, and keep asking those big questions.

앱 다운로드

유창하게 말하는 데 필요한 모든 것

AI 발음 평가문장마다 점수 확인
IPA 발음 연습모든 소리 마스터
단어 학습나만의 단어장 만들기
단어 게임놀면서 외우기
FOR YOU

추천 영상

이 레슨의 어휘와 말하기 포인트

이 말하기 레슨은 영상 “Episode 72 : Neuralk - AI for Structured Data”을(를) 바탕으로 합니다. 가장 자주 반복되는 단어는 다음과 같습니다: model, product, Neuralc, structured, exactly. 이 영상에는 섀도잉할 문장 161개와 단어 1305개가 있습니다. 말하는 구간의 길이는 7:33입니다. 화자는 분당 약 173단어로, 일상 대화에 가까운 자연스러운 속도로 말합니다. 단어의 82%가 영어에서 가장 많이 쓰이는 3,000단어에 속합니다. 나머지는 연습 전에 미리 확인해 두세요.

이 영상의 핵심 어휘

영상에 나오는 익혀 둘 만한 단어 15개를 발음, 뜻과 함께 정리했습니다.

단어발음뜻
analyze 동사/ˈæn.əˌlaɪz/분석하다
database 명사/ˈdeɪtəˌbeɪs/데이터베이스
predict 동사/pɹɪˈdɪkt/예언하다, 예측하다
attribute 명사/ˈæt.ɹɪˌbjut/속성, 특질
listener 명사/ˈlɪs(ə)nɚ/경청자, 청취자
solve 동사/sɒlv/해결하다
commerce 명사/ˈkɑ.mɚs/무역, 교역
rely 동사/ɹɪˈlaɪ/의지(依支)하다
shoe 명사/ˈʃu/신발, 신
dive 동사/daɪv/잠수하다
analytics 명사/ˌæn.əˈlɪt.ɪks/해석학, 애널리틱스
topic 명사/ˈtɑpɪk/화제, 주제
vast 형용사/væst/광대하다, 거대하다
audio 형용사/ˈɔ.di.oʊ/오디오
impressive 형용사/ɪmˈpɹɛsɪv/인상적

주의할 발음

화자는 they're, don't, I'm 같은 축약형과 약화된 형태를 25번 사용합니다. 들리는 대로 짧게 발음하세요.

  • “sh”와 “zh” 소리: benchmark /ˈbɛn(t)ʃmɑːk/, shoe /ˈʃu/, reputation /ˌɹɛpjʊˈteɪʃən/, chef /ʃɛf/, prediction /pɹɪˈdɪkʃən/
  • 긴 단어 — 강세 위치에 주의: analytics /ˌæn.əˈlɪt.ɪks/, reputation /ˌɹɛpjʊˈteɪʃən/, developer /dɪˈvɛləpɚ/, effectiveness /ɪˈfɛk.tɪv.nɪs/, accordingly /əˈkɔɹ.dɪŋ.li/

한국어 화자가 어려워하는 소리:

  • /f/ — ㅍ(/p/)으로 바꾸지 말고 윗니를 아랫입술에 대기: chef /ʃɛf/, preference /ˈpɹɛf(ə)ɹ(ə)ns/, effectiveness /ɪˈfɛk.tɪv.nɪs/, flavor /ˈfleɪvə/, notification /ˌnoʊtɪfɪˈkeɪʃn̩/
  • /v/ — ㅂ(/b/)과 구별하기: solve /sɒlv/, dive /daɪv/, vast /væst/, impressive /ɪmˈpɹɛsɪv/, discovery /dɪˈskʌv.ɹi/
  • /z/ — ㅈ이 아니라 성대를 울리는 /s/: analyze /ˈæn.əˌlaɪz/, laser /ˈleɪzɚ/, browse /bɹaʊz/, categorize /ˈkætɪɡəˌɹaɪz/, personalize /ˈpɜː.sə.nə.laɪz/

이 영상으로 연습하는 방법

  1. 먼저 말하지 않고 영상을 끝까지 듣고 모르는 단어를 적어 둡니다.
  2. 0.75배속으로 한 문장씩 섀도잉을 시작하고, 익숙해지면 보통 속도로 돌아갑니다.
  3. 자신의 목소리를 녹음해 원본과 비교하고, analyze, database, predict 같은 단어에 특히 주의합니다.

쉐도잉이란? 영어 실력을 빠르게 키우는 과학적 방법

쉐도잉(Shadowing)은 원래 전문 통역사 훈련을 위해 개발된 언어 학습 기법으로, 다언어 학자인 Dr. Alexander Arguelles에 의해 대중화된 방법입니다. 핵심 원리는 간단하지만 매우 강력합니다: 원어민의 영어를 들으면서 1~2초의 짧은 지연으로 즉시 소리 내어 따라 말하는 것——마치 '그림자(shadow)'처럼 화자를 따라가는 것입니다. 문법 공부나 수동적인 청취와 달리, 쉐도잉은 뇌와 입 근육이 동시에 실시간으로 영어를 처리하고 재현하도록 훈련합니다. 연구에 따르면 이 방법은 발음 정확도, 억양, 리듬, 연음, 청취력, 말하기 유창성을 크게 향상시킵니다. IELTS 스피킹 준비와 자연스러운 영어 소통을 원하는 분들에게 특히 효과적입니다.

섀도잉 방법: 단계별 전체 가이드 읽기 →