跟讀練習: Run AI Models (LLMs) from USB Drive | No Filter, Fully Offline💀 - 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。
102 句
如果句子過短或過長，請點選 Edit 進行調整。
1
This little USB drive has a full AI model on it.
2
No internet, no account, and no filter.
3
I just plugged it in, double clicked once, and it's running.
4
It's all on my own hardware.
5
So nothing goes to the cloud, and your chats stay right on this drive.
6
And since everything runs locally on your machine, there are no company filters.
7
Just ask whatever you want and get straight answers, without the usual refusals or lectures.
8
And it's not just chat.
9
The same app can generate images, turn your voice into text, and read answers back out loud, all offline.
10
And you are not stuck on one operating system.
11
It works on Windows, Mac, and Linux, and it uses your graphics card automatically, whether that's NVIDIA, AMD, Intel, or Apple Silicon.
12
So this is Portable Local Studio.
13
It's completely free and open source, and once it's on this drive, it's all yours.
14
I put a lot of hours into building this, so if you find it useful, please consider subscribing to the channel.
15
It really helps me a lot.
16
Alright, now I'll show you what it can do first, and then I'll show you the installation.
17
Let's start with chat.
18
Let me turn off my Wi -Fi first.
19
Here, we have the option to select a model.
20
Once you select it, it loads instantly.
21
We also have the option to attach a file, enable Web Search, and turn on thinking.
22
Web search needs the internet, so I'll leave that off for now, but everything else runs completely offline.
23
Let me turn on thinking and ask it something.
24
You can see it's thinking first, working through the problem, and then it gives the answer.
25
And remember, Wi -Fi is off.
26
This is all running from the USB drive.
27
Now let me attach a file and ask it a question. And there you go.
28
It reads the file and gives me an answer, without anything leaving my computer.
29
And here's the no filter part.
30
Most AI apps would refuse a question like this, but it just answers, with no lecture at all.
31
Now let's move to image generation.
32
I'll switch over to the image workspace.
33
And just like with chat, I can pick an image model here.
34
Once I select one, it starts loading.
35
Once the model is loaded, I can just type in what I want to see.
36
Let's try something crazy. And there you go.
37
The image is generated right here on my machine with no internet at all.
38
And if you want more control, you can change the steps, sampler and seed.
39
If you're new, just leave the defaults.
40
They work well.
41
And everything you create is saved in your history, together with the prompt and settings, all on the USB drive.
42
Next, speech to text.
43
Just like before, I pick a model here and it loads up.
44
Now I can record my voice right here, or drop in an audio file, like a voice note or a lecture.
45
Just hit transcribe, and it transcribes your audio locally in just a few seconds.
46
And there it is, my voice turned into text.
47
You can copy it or save it as a file.
48
And you can also pick the spoken language, so it works for more than just English.
49
Now let's try text -to -speech.
50
First I'll select a voice model, and it loads in just a few seconds.
51
Now I'll paste in some text and pick a voice.
52
There are male and female voices, with American and British accents.
53
Once you've picked one, just tap
54
Speak, and it will turn your written text into actual speech. And there it is.
55
A natural sounding voice, generated right on my machine, with no internet and no voice subscription.
56
Alright, now let me show you how to set it all up on your own USB drive.
57
I'll do it on Windows, but the process is almost the same on Mac and Linux.
58
First, you need a USB drive.
59
I recommend at least 32GB, and USB 3 .0 or fast of a good speed.
60
Before anything else, format the drive as exfat.
61
That way it works on Windows, Mac and Linux.
62
Next, head over to the GitHub page.
63
I'll leave the link in the description.
64
Click the green code button, then download zip and extract it straight onto your USB drive.
65
Now let's launch it.
66
On Windows, just double click the windows .bat file.
67
The first time it needs an internet connection to set things up.
68
Everything stays on the USB drive.
69
So let's wait for it to finish. And there you go.
70
Our local studio opened in the browser automatically.
71
Now let's head over to the models page to download AI models.
72
Here you can browse image, text, speech, and voice models.
73
It also recommends models that best fit your hardware.
74
Want no filter answers? just tap the uncensored filter.
75
Once you find one you like just hit download and it saves straight onto the USB drive.
76
Once it's downloaded, you'll see it right here in chat.
77
Just select it and you can start chatting.
78
You can do the same for image generation.
79
I recommend Epic Realism, it gives really realistic results, and it has no built -in filter either.
80
The model is downloaded, so let's give it a try. And there it is.
81
Image generation is working perfectly, right from the USB drive.
82
Now let's take this same USB drive and plug it into a Mac.
83
On Mac, open a terminal in the folder and run the Mac script.
84
The first time on a Mac, it needs the internet for a minute to set up.
85
The models are already on the drive, so there's no need to download them again. And there it is.
86
The studio opens in the browser with all the same models.
87
Even our chats and images from Windows are still here.
88
Let's quickly test text generation with the Gemma 4 model we downloaded on Windows.
89
And yes, it's running on a Mac as well.
90
Now let's take the same USB drive and plug it into a Linux machine.
91
On Linux, open a terminal and run the Linux script.
92
It needs internet for the first setup, but your models are already on the drive. And there it is.
93
The studio opens in the browser with all the same models and again our chats and images are all still here.
94
Let's quickly test text generation with the same Gemma 4 model.
95
And yes, it's running on Linux too.
96
One USB drive, three operating systems.
97
So that's it.
98
A full AI studio on a single USB drive.
99
Chat, images, voice, all running offline, on Windows, Mac and Linux.
100
If you found this video helpful, please consider subscribing.
101
It really helps the channel.
102
Thanks for watching, and I'll see you in the next one.
FOR YOU
推薦影片
B21
76 句
Nvidia's New Laptops - RTX Spark
Daily ConversationWorking Adults
B21
329 句
Laziest Ways to Make Money with AI (For Beginners)
Daily ConversationWorking Adults
B21
305 句
The Wall Outlet Computer
Daily ConversationWorking Adults
B23
43 句
Translating Claude’s thoughts into language
Daily ConversationUniversity StudentsWorking Adults
B22
193 句
How I would learn English if I started now - 2024
Daily ConversationWorking Adults
B21
123 句
Google Just Dropped NEW AI Updates!🔥
Daily ConversationWorking Adults
B21
265 句
Neuralink’s Newest Brain Chip Isn’t What You Think
Daily ConversationWorking Adults
B21
331 句
Dear Higgsfield, I reverse engineered you in 12 hours
Daily ConversationUniversity StudentsWorking Adults
B2385 句
11 AI Breakthroughs You Didn’t Know Existed
Daily ConversationWorking Adults
B21
160 句
How AI Is Changing the Working World
Daily ConversationWorking Adults
B21
124 句
How to Build AI ChatBot with Custom Knowledge Base in 10 mins
Daily ConversationWorking Adults
B21
24 句
DeepSeek R1 Explained in 2 Minutes
Daily ConversationUniversity StudentsWorking Adults
本課的詞彙與口語要點
這節口語課以影片“Run AI Models (LLMs) from USB Drive”為素材。 影片中反覆出現的詞有：drive, model, Usb, image, voice。 這段影片共有 102 個句子、1140 個單詞可供跟讀。 講話部分時長為 6:59。 說話人語速自然，每分鐘約 163 個詞，接近日常對話。 83% 的單詞屬於英語最常用的 3,000 詞，其餘的詞建議在練習前先學一下。
影片中的重點詞彙
影片中 15 個值得學習的單詞，附發音和釋義：
| 單詞 | 發音 | 釋義 |
|---|---|---|
| filter 名詞 | /ˈfɪltɚ/ | 過濾器 /過濾器, 濾過器 /濾過器 |
| select 動詞 | /sɪˈlɛkt/ | 選 /選, 挑選 /挑選 |
| lecture 名詞 | /ˈlɛk.t͡ʃə/ | 演講 /演講 |
| plug 名詞 | /plʌɡ/ | 插頭 /插頭 |
| offline 形容詞 | /ɔfˈlaɪn/ | 離線 /離線, 離線 /離線 |
| audio 形容詞 | /ˈɔ.di.oʊ/ | 音訊 /音訊, 伴音 |
| script 名詞 | /skɹɪpt/ | 文章 |
| terminal 名詞 | /ˈtɚmɪnəl/ | 航站, 航廈 /航廈 |
| hardware 名詞 | /ˈhɑɹdˌwɛɹ/ | 硬體, 硬體 /硬體 |
| subscribe 動詞 | /səbˈskɹaɪb/ | 簽署 /簽署 |
| attach 動詞 | /əˈtæt͡ʃ/ | 附上, 附 |
| format 名詞 | /ˈfɔːɹ.mæt/ | 格式, 開本 /開本 |
| cloud 名詞 | /ˈklaʊ̯d/ | йүн, 雲 /雲 |
| helpful 形容詞 | /ˈhɛlp.fl̩/ | 有幫助 /有幫助 |
| refuse 動詞 | /ɹɪˈfjuːz/ | 拒絕 /拒絕, 回絕 /回絕 |
影片中出現的短語動詞
| 單詞 | 釋義 |
|---|---|
| turn off 動詞 | 關 /關, 關掉 /關掉 |
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：
- It's all on my own hardware.
- I'll switch over to the image workspace.
- If you're new, just leave the defaults.
- I'll leave the link in the description.
需要注意的發音
說話人用了 12 次縮略和弱讀形式，例如 I'll, you'll, you're。請按聽到的簡短形式來說。
- “sh” 和 “zh” 音: installation /ˌɪn.stəˈleɪ.ʃən/, subscription /səbˈskɹɪpʃən/
中文母語者容易讀錯的音：
- 詞尾子音 — 要讀清楚，後面不要加母音: select /sɪˈlɛkt/, generate /ˈd͡ʒɛn.ə.ɹeɪt/, plug /plʌɡ/, script /skɹɪpt/, subscribe /səbˈskɹaɪb/
- /r/ 和 /l/ 的區別: realistic /ˌɹɪəˈlɪs.tɪk/, refusal /ɹɪˈfjuːzl̩/
- /æ/ — 嘴張大，不要讀成 /e/: attach /əˈtæt͡ʃ/, format /ˈfɔːɹ.mæt/, graphics /ˈɡɹæfɪks/, accent /ækˈsɛnt/, extract /ˈɛkstɹækt/
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。
- 先用 0.75 倍速逐句跟讀，熟練之後再回到正常速度。
- 錄下自己的聲音並與原聲對比，特別注意 filter, select, lecture 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。