Unit 18
Hey Siri!
新知思想/趣味科普
New Knowledge/Fun Science
BRAIN STORM
Share your ideas with your teacher and try to make sentences.
Are you familiar with Siri? What do you think is the purpose of Siri?
VOCABULARY
Repeat these sentences after your teacher.
1. speech recognition (phr.) 語音辨識
Siri identifies your voice by speech recognition.
Siri通過語音識別來識別您的聲音。
2. corresponding (adj.) 相應的;對等的
We need to design this with corresponding colors.
我們需要使用相應的顏色進行設計。
3. cater to (idi.) 滿足… 願望或需求
Do you think Siri can cater to a foreign voice?
您認為Siri可以迎合外來聲音嗎？
4. transcribed (n.) 轉錄，轉寫
John transcribed the article for you.
約翰為您抄寫了這篇文章。
5. comprised of (idi.) 由… 組成
Smart phones are now comprised of Ai assistants.
智能手機現在由Ai助手組成。
6. Natural Language Processing (phr.) 自然語言處理
Siri has Natural Language Processing.
Siri具有自然語言處理功能。
7. algorithms (n.) 演算法
YouTube algorithms led me to your channel.
YouTube算法引導我進入了您的頻道。
8. mammoth (adj.) 巨大的
This will be a mammoth undertaking.
這將是一項艱鉅的任務。
Dialogue
Read the dialogue aloud with your teacher.
Hey, Siri. Set an alarm for 6:30 AM tomorrow morning.
嘿，Siri。 定一個明早六點半的鬧鐘。
Okay. I set an alarm for 6:30 AM.
好的。 我已把鬧鐘定在明早六點半。
Boy, speech recognition technology sure has come a long way. Have you ever
stopped to think about how it works?
厲害了，語音辨識技術已經取得了這麼大的進步。 你有沒有想過它是什麼原理呢？
Actually, I have. When you trigger Siri by speaking, your speech is converted from audio to its corresponding textual form in the back-end of the system.
我還真想過。 當你通過說話來喚醒Siri的時候，你說的話會在系統後端從語音形式轉換成相應的文本形式。
I see. Developing such software must have been extremely challenging.
我明白了。 開發這樣的軟體一定是個巨大的挑戰。
And time-consuming. Siri is able to cater to various accents because engineers train machine learning models on large, transcribed datasets comprised of voice samples from large groups of people.
而且非常耗時。 Siri能夠識別各種各樣的口音是因為工程師們訓練了機器學習模型，這是基於龐大的人群語音轉寫樣本的數據集完成的。
That’s fascinating.
真有意思。
Yeah, but what I find fascinating is that Apple servers run Natural Language
Processing, or NLP, algorithms to understand the intent of what the user is trying to say.
是啊，但我覺得特別吸引人的是蘋果的伺服器會運行自然語言處理技術，也就是NLP，這是一種可以理解說話人意圖的演算法
Which enables NLP engines to distinguish exactly what someone is saying?
也就是能讓NLP引擎分辨出某個人到底在說什麼嗎？
You got it. But doing so is a mammoth task because of the many different ways you can say one thing.
你說對了。 但實現這項技術其實是個非常龐大的任務，因為有太多不同的方式可以表達同一件事了。
I imagine contextual understanding must be another major aspect of Siri’s machine learning technology then?
那我猜語境理解就是Siri機器學習技術的另一個重要方面了吧？
Sure is. That’s why more and more we’re able to speak to Siri like any other human. Hey Siri, play “Yellow Submarine” by the Beatles.
沒錯。 這就是為什麼和Siri對話越來越像和人對話了。 嘿，Siri。 播放披頭士的Yellow Submarine。
Grammar
Clause with “what…” to emphasize
In this lesson, the speaker says “What I find fascinating is that…” This is an example of a cleft sentence. It can be said another way: “The thing that I find fascinating is…” Cleft sentences are very common. What they do is introduce and emphasize what you’re about to say and they can often help to contrast it with other things or even what somebody else said.
For example: A: I don’t really like Siri’s voice. B: Really? What I dislike about Siri is that she never understands me. (Here, “she never understands me” is emphasized and this is being contrasted with A’s statement.)
Here are a few patterns:
1. What clause + is + noun phrase/wh-clause/how-clause
What I need is an ice cold beer.
What I don’t get is where all this money is going. It just all seems to disappear!
2. What clause + is + bare infinitive
What you should do is go to bed.
3. What clause + is + (that) + subject + verb
What I found ridiculous was that the price was double for tourists.
1. What I don’t understand is how he could stay in such an abusive relationship.
2. What you need to do is consult an accountant for better financial advice.
3. What I find confusing is this grammatical structure.
DISCUSSION
Make a conversation with your teacher.
1.Can you give other examples of ai assistants that you can find on smart phones?
