跟读练习: How to Make Viral Stickman Animation With Fable 5.2 (Full Breakdown) - 通过视频学习英语口语

选择练习模式

每种模式都会打开全屏练习室。

209 句

如果句子过短或过长，请点击 Edit 进行调整。

Automated shorts like this are completely taking over YouTube, pulling millions of views and building massive channels in weeks.

这类自动化生成的短视频正在彻底占领YouTube，短短几周内就能获得数百万播放量，打造出规模庞大的频道。

Accounts posting this exact format are pulling thousands of dollars every single month,

采用这种固定格式发布内容的账号每个月都能赚几千美元。

purely from short -sized revenue and high -paying sponsorship deals without ever showing their faces.

完全依靠短视频收入和高额赞助合作，全程无需露脸。

Now, most people think you need either Blender or Cloud Code to build these,

现在，大多数人认为要制作这些东西，你要么得用 Blender，要么得用 Cloud Code。

but today I'm breaking down the exact workload to make these viral animations from scratch using just a chat window in Cloud.

但今天我要详细拆解，仅通过云端的一个聊天窗口从零开始制作这些爆款动画的具体工作量。

Hi, I'm Karima, and welcome back to my creative space.

大家好，我是卡里玛，欢迎回到我的创意空间。

By the end of this video, you'll have the only step -by -step workflow you need to build and monetize your viral AI animations in minutes,

看完这个视频，你就能掌握一套循序渐进的操作流程，只需几分钟就能制作出爆款AI动画并实现变现。

even if you have zero editing or 3D experience.

即使你完全没有剪辑或3D制作经验。

On the surface, these stickman videos look like simple little animations people post for laughs.

表面上看，这些火柴人视频就像是大家发出来搞笑的简单小动画。

But here is what is actually happening behind the scenes.

但幕后的实际情况却是这样的。

Creators are building faceless channels around this exact format and getting paid in three different ways at once.

创作者们正围绕这种固定模式打造不露脸的账号，同时通过三种不同方式获得收益。

That's YouTube Shorts ad revenue, big sponsorships from AI companies that are actively paying creators who post animated stories,

这包括YouTube Shorts的广告收入，以及来自人工智能公司的大额赞助——这些公司正在积极向发布动画故事的创作者支付费用。

and three, that's making these high -retention videos for other creators and brands.

第三点，就是为其他创作者和品牌制作这些高留存率的视频。

Before filming this breakdown, I watched a ton of tutorials on YouTube and tested the workflows everyone else is teaching.

拍摄这期解析视频之前，我在YouTube上看了大量教程，还测试了其他人分享的所有工作流程。

Honestly, they were slow, confusing, and burned through my money.

说实话，他们效率低下、流程混乱，还花掉了我不少钱。

So here's why this tutorial is the only one you need.

所以这就是为什么这份教程是你唯一需要的。

You don't need Blender, you don't need ClockCode, and you definitely don't need any editing software.

你不需要Blender，不需要ClockCode，也完全不需要任何剪辑软件。

All we're using today is the regular Claude chat window.

我们今天用的就是普通的 Claude 聊天窗口。

And yes, it is actually that simple.

没错，实际上就是这么简单。

Now here's a twist that makes this whole workflow possible.

现在有一个关键转折，让整个工作流程得以实现。

Claude Fable 5 .2 just dropped and everyone online has been going crazy over what it can do.

Claude 3.5 Sonnet 刚刚发布，网上所有人都在为它的强大功能疯狂。

Fable 5 .2 is Claude's newest model built specifically for visual understanding, animations, 3D spatial logic, and building full videos with sound.

Fable 5.2 是 Claude 的最新模型，专为视觉理解、动画、3D 空间逻辑以及生成带音频的完整视频而打造。

Even better, today you don't even need to know how to write prompts.

更棒的是，如今你甚至不需要懂得如何编写提示词。

There are zero prompts you have to write in this entire workflow

在整个工作流程中，你完全不需要编写任何提示词。

and i'm honestly still shocked at how simple i managed to make

说实话，我到现在还很震惊，我居然能把事情做得这么简单。

this along the way i'll show you the exact ready cost for every test

过程中我会给你展示每一项检测的实际准备成本。

make sure you stick around until the very end

一定要看到最后哦

because i'm giving away the custom ai skills i built

因为我要把自己开发的自定义AI技能免费分享出去

that automates this whole pipeline completely for free grab your drink open your laptop

它能完全免费地将整个流程自动化，拿上你的饮品，打开你的笔记本电脑。

and let's get into it before

那我们就开始吧。

we jump into a step -by -step build in just a minute

我们马上就开始一步步搭建。

you need to understand why beginners quit before they even start i watched a bunch of youtube tutorials

你得搞清楚为什么新手还没开始就放弃了，我之前看了一堆YouTube教程。

and found out there are three main ways people make stickman animations like this

发现人们制作这类火柴人动画主要有三种方法

and i tested all three of them so

而且我这三个都试过了，所以

that you don't waste your own time

不要浪费你自己的时间

and money way number one is connecting claude to blender

最有效的方法之一是将 Claude 接入 Blender。

so claude builds the 3d animations inside blender for you i tried working this way first

所以Claude会在Blender里帮你制作3D动画，我一开始就是尝试用这种方式工作的

and here is what nobody tells you

而这就是没人会告诉你的事

if you don't already know blender the interface is super confusing

如果你还不熟悉Blender的话，它的界面会让人特别摸不着头脑。

and it just takes so much time for claw to build

而且搭建Claw真的要花很多时间。

and export each scene even worse

导出每个场景的效果甚至更差

it eats up an insane amount of tokens building a single short video

生成一段短短视频就要消耗掉极其大量的token

this way can easily burn through 200 000 to 300 000 claw tokens

这种方式很容易消耗掉20万到30万个Claw代币。

which hits your usage limits very fast

这会让你很快就达到使用上限

and take a look at the result i got i mean it looks okay

你看看我得到的结果，我觉得看起来还不错。

but it's still not what i wanted

但这仍然不是我想要的。

so next i tested method number two

接下来我测试了第二种方法

that's building the animation using cloud code only

这完全是通过云代码来构建动画的

now the talking cost here was relatively okay compared to the blender setup but

现在和Blender的设置相比，这里的通话成本还算可以，但

when i opened the video it made

当我打开视频时，它就开始播放了

i was still not satisfied enough look at what clock code gave me

我还是不太满意，你看看Clock Code给我的结果。

i mean the animations look kind of basic i expected way more from it

我的意思是这些动画看起来有点简陋，我原本对它的期望要高得多。

and the video outputs they just don't look good enough in my opinion

而且在我看来，这些视频输出的效果实在不够好。

because both of those methods were either too slow too expensive

因为这两种方法要么速度太慢，要么成本太高

or gave mid results i decided to test a third way that's using the new fable 5 .2 model inside the regular

或者效果一般，我决定测试第三种方法，就是在常规环境里使用新的Fable 5.2模型。

Cloud Chat and connecting it to HiggsField MCP.

Cloud Chat 及其与 HiggsField MCP 的连接配置。

Quick definition for my beginners, if you haven't heard of MCP before, don't be worried at all.

给新手们简单解释一下，如果你们之前没听说过MCP，完全不用担心。

It just acts like a bridge that lets

它的作用就像一座桥梁，能够让

Cloud Chat generate image and videos using HiggsField models automatically behind the scenes.

Cloud Chat 会在后台自动使用 HiggsField 模型生成图像和视频。

And if you've never opened the HiggsField website and felt confused by all the different tools and buttons in the top bar,

如果你从未打开过 HiggsField 网站，也未曾因顶部栏里各式各样的工具和按钮感到困惑，

I have great news for you.

我有个好消息要告诉你。

With this workflow, you don't have to use the website at all.

使用这个工作流程，你完全不需要访问该网站。

The only time when you actually have to open it is at the very start to sign up

你唯一需要打开它的时候就是最开始注册的时候。

and connect Hicksfield MCP to Cloud.

并将 Hicksfield MCP 连接到云端。

I'll leave the link in the description below.

我会把链接放在下方的简介里。

Once you do that 30 -second setup, you can literally forget about Hicksfield, and now every single thing we do from here happens inside a normal

完成那30秒的设置后，你完全可以不用再管Hicksfield了，从现在开始，我们所有的操作都在一个普通的环境里进行。

Cloud chat window with Fable 5 .2.

与Fable 5 .2的云聊天窗口。

Okay, the prep work is officially left behind, and now we get to the part you've all been waiting for, that's generating the actual animation.

好的，准备工作已经正式完成，现在我们来到了大家一直期待的环节，也就是生成实际的动画。

Now don't bother taking notes because I put the whole workflow

现在不用费心记笔记，因为我已经把整个工作流程都整理好了

and my custom AI skill in the free guide at the end of the video.

以及视频末尾免费指南中我自制的AI技能。

Now let's jump to step 1.

现在我们直接进入第一步。

Now first I open

现在我先打开

Cloud, select Fable 5 .2 from the model drop down

在云端界面，从模型下拉菜单中选择 Fable 5.2。

and set the effort to high and drop in my custom AI animation skill.

然后将难度设为高，再加入我自定义的AI动画技能。

Now people are going to ask, where did I get this skill from?

现在大家肯定会问，我这本事是从哪学来的？

So after spending days testing this format and making a bunch of these animated videos myself, To make your life easier, I collected all my rules,

所以我花了好几天测试这种格式，还自己做了一大堆这类动画视频，为了帮大家省点事，我整理出了所有的制作规则。

layout structures, and editing instructions into a custom AI skill file, and I'm giving it away for free at the end of the video,

排版结构和编辑指令整合到一个自定义AI技能文件中，我会在视频结尾免费分享这个文件。

so make sure to stick until the very end.

所以一定要坚持看到最后。

Next, I read a short funny script about Leonardo da Vinci

接下来，我读了一篇关于列奥纳多·达·芬奇的有趣短剧本。

getting hired in the Silicon Valley where he rides into a modern tech campus.

他成功入职硅谷，每天通勤前往现代化的科技园区上班。

On a horse, holding his parchment scroll resume, he meets a tech recruiter and tests out their high -tech AI lab.

他骑在马上，手持羊皮卷简历，与一名科技招聘人员会面，并体验了对方的高科技人工智能实验室。

This script is what the video model is going to make the animation about, so your script can be literally on any other topic.

这个脚本就是视频模型要用来制作动画的依据，所以你的脚本主题完全可以是其他任何内容。

After that, I type my first brief in a very simple way.

之后，我用非常简单的方式输入了我的第一份简报。

So I just say, I'll give you the script, and you will generate and assemble the video.

那我就直接说，我给你脚本，你来生成并拼接成视频。

Run a test generation using Gemini Omni Flash 1 .2 at 1080p with this script about DaVinci.

使用 Gemini Omni Flash 1.2 以 1080p 分辨率运行测试生成，脚本为这份关于 DaVinci 的内容。

Generate every scene in order.

按顺序生成每一个场景。

Start by making character sheets in the Sigma style first.

首先先以 Sigma 风格制作角色卡。

And make sure the opening hook is super catchy.

还要确保开头的钩子足够吸引人。

Now pay close attention to the video generation model I chose.

现在请仔细关注我选择的视频生成模型。

That's Gemini Omni Flash 1 .2.

那是 Gemini Omni Flash 1.2。

If you follow my channel, you know I usually use CDN 2 .5 for hyper -realistic films and commercials.

关注我频道的观众都知道，我通常会用CDN 2.5来拍摄超写实风格的电影和广告。

But as someone who has generated hundreds of AI videos over the last year,

但作为过去一年里制作了数百条AI视频的人，

I always tell you guys that there is no single best model for everything.

我一直跟大家说，不存在一个能胜任所有任务的最佳模型。

You pick the right paintbrush for the right job.

你要为不同的工作挑选合适的画笔。

After testing multiple video models on this specific Stickman style, I realized that Gemini Omni Flash is hands down the best model for this format.

在针对这种特定火柴人风格测试了多款视频模型后，我发现Gemini Omni Flash无疑是最适合这种格式的模型。

It nails the contrast between a flat 2D cartoon Stickman and a 3D environment without making the character look weird,

它完美呈现了平面二维卡通火柴人与三维环境之间的对比，同时又不会让角色看起来很怪异。

and here is the massive budget hack to save your credits it is actually incredibly cheap.

接下来要分享一个超实用的预算技巧，能帮你省下学分，而且成本其实非常低。

Take a look at the cost comparison right here.

请看一下这里的成本对比。

Generating a 15 second sequence with Gemini Omni Flash costs only 48 credits in 1080p and just 33 credits in 720p.

使用 Gemini Omni Flash 生成一段15秒的视频序列，1080p分辨率仅需48积分，720p分辨率仅需33积分。

100

Whereas, running that same duration on Seedense costs 240 and 105 credits relatively.

而在 Seedense 上运行相同时长的任务，分别需要消耗 240 和 105 个积分。

101

That is a massive price difference.

这差价也太大了。

102

Also, if you have the opportunity to spend a little more credits, Gemini Omni flash is available in 4k

另外，如果您有机会多花一点积分，Gemini Omni 闪光灯支持 4K 分辨率。

103

but honestly I think 1080p works just fine for video formats like this

不过说实话，我觉得对于这类视频格式来说，1080p的画质已经完全够用了

104

so before generating any video Fable creates a three panel 2d character sheet for Leonardo da Vinci

因此，在生成任何视频之前，Fable会为列奥纳多·达·芬奇创建一个包含三个面板的2D角色设定图。

105

and look at how clean

看看这多干净

106

that sheet is he has the round white egghead symbol dot eyes his long gray beard and the exact

那张图里的他有着圆圆的白色光头符号、点状眼睛、长长的灰色胡须，而且完全一模一样

107

Renaissance hat and tunic okay after that cloud finished working

那朵云处理完之后，文艺复兴风格的帽子和束腰外衣就没问题了

108

and dropped the video file right here let's open it up

然后把视频文件拖到这里了，我们打开看看

109

and check it out when Leonardo da Vinci applied for an engineering job in Silicon Valley.

还有，你一定要看看列奥纳多·达·芬奇申请硅谷工程师职位的那段内容。

110

At first glance, I really like what it does for me.

乍一看，我真的很喜欢它给我带来的效果。

111

So the humor lands, the pacing is super fast, and seeing a 2D character like da Vinci walking around a modern tech office looks super cool.

笑点很到位，节奏非常快，而且看到达芬奇这样的2D角色在现代科技办公室里走来走去的样子简直帅爆了。

112

But listen up, I noticed three specific things that actually bother me.

但听我说，我注意到有三件具体的事确实让我很困扰。

113

First, that's the art style.

首先，这就是它的艺术风格。

114

It shifts slightly between a couple of scenes instead of staying locked to one exact look.

它在几个场景之间会有细微的变化，而不是始终保持完全一致的外观。

115

Second, it didn't check carefully enough for the AI slot, Like, in the photo scene with the recruiter,

其次，它对AI生成片段的检查不够仔细，比如在有招聘人员的那个照片场景里，

116

a random third hand literally just appears on him.

莫名其妙就凭空多出来了第三只手搭在他身上。

117

Wait, what?

等等，你说什么？

118

What the hell is this?

这到底是什么鬼东西？

119

And third, right at the very beginning, it isn't obvious enough in the opening hook that he is in the Silicon Valley.

第三点，在最开头的引入部分，并没有足够明确地交代他身处硅谷。

120

Now, the best part is that I don't have to start over to fix this.

现在最棒的是，我不用从头再来就能解决这个问题。

121

I just typed a two -sentence request straight back to Claude.

我刚直接给 Claude 输入了一个只有两句话的请求。

122

Claude went back to work, replaced the flawed clips, put the edit back together, and gave me the new file.

克劳德回去继续工作，替换了有问题的片段，重新拼接好剪辑，然后把新文件发给了我。

123

When Leonardo da Vinci applied for an engineering job in Silicon Valley.. okay wow look at this version right here

要是列奥纳多·达·芬奇去硅谷应聘工程师职位……哇，快看这个版本

124

i love everything about how it turned out first look at how locked

我太喜欢最终呈现出来的所有效果了，你先看看这质感有多扎实

125

or character sheet is across every single scene

或者说角色设定贯穿了每一个场景

126

and listen to the audio narration it's just

然后听一下音频解说，简直就是

127

so cool second look at the background details in the opening hook

太酷了，再仔细看一下开场钩子部分的背景细节

128

you immediately see him riding his horse through a sunny silicon valley parking lot with sleek glass buildings

你立刻看到他骑着马穿过阳光明媚的硅谷停车场，四周是造型流畅的玻璃大楼。

129

palm trees

棕榈树

130

and wife's self -driving cars the recruiter in the blue hoodie okay he's sitting at his desk with his laptop

还有妻子的自动驾驶汽车，穿蓝色连帽衫的招聘人员，好的，他正坐在桌前用着笔记本电脑

131

and ice drink and

132

when da vinci stands outside the glass lap door you can clearly see the glowing red card reader

133

so the white drone hanging from the ceiling

134

and the robotic arm on the workbench side

135

now let's check the scoreboard for step one building

136

this entire 2d animated short with fable 5 .2 including the character sheets all the video scenes the iteration

137

and the final assembled edit cost me roughly 336 credits quick thing before we move to step two,

138

only 10 % of you watching right now are actually subscribed.

139

So to stay up to date with the latest trends and be the first to watch my tutorials,

140

hit that subscribe button and notification bell right now.

141

Also, don't bother writing any of these steps down.

142

The full setup is in the free guide

143

and i'm sharing it with you at the end of the video now

144

that 2d stickman look is what everyone on youtube shorts is posting right now

145

but in step two i want to test how far

146

we could push fable 5 .2 by taking this exact same story

147

and leveling it up into a full 3d animation

148

now step two is how you make your channel stand out

149

from every other creator in your niche having made hundreds of viral videos

150

that brought me over a million followers in just six months.

151

Here is the lesson that changed everything for me.

152

Once a format is proven to go viral on the algorithm, you don't need to reinvent the wheel.

153

You just upgrade the idea by 20 % so nobody can scroll past yours.

154

Now most stickman channels make videos in 2D

155

and since Fable 5 .2 was built specifically with 3D spatial reasoning and animation in mind,

156

I thought what if we keep our exact same funny stickman proportions, the round white head, the simple eyes, the thin sick legs

157

but turned the whole video into a stylized 3d animation normally

158

turning a 2d project into 3d would mean starting over from zero

159

but with claude and hicksville mcp watch how easy this is

160

so first i tell claude to regenerate the 2d character sheets of her stickman leonardo da vinci into 3d

161

so we have the same portrait the same full body view with the thin sick legs side profile

162

but now with 3d volume

163

and shading then i stayed right inside clod with fable 5 .2 set it to hot effort attached

164

that 3d character sheet alongside my script

165

and my custom skill and typed one simple sentence with these assets the exact same story

166

but let's make the full video in 3d animation now

167

while we wait here's what people get wrong about ai they hear clod can build a whole video from scratch

168

and think the creator doesn't matter anymore.

169

It's the exact opposite.

170

Yeah, it is just the digital paintbrush while you're still the director calling the shots.

171

Okay, Claude just replied with our 3D video file and honestly, I love how this turned out.

172

Let's watch it.

173

Look at how much more interesting and dynamic the 3D style feels.

174

Like, every single detail from our 3D character sheet is preserved across the scenes.

175

Even better, look at how he fits into backgrounds now.

176

I don't know, I personally love watching these kind of videos especially

177

if they cover interesting sci -pop topics

178

and again in case you're testing this personally

179

and you feel like you want to change something just write your iteration to claude

180

and ask to regenerate the video for you take a look at our total cost for step two generating

181

and assembling this entire 3d version with fable 5 .2 cost me just 290 credits

182

so after testing all three methods what's the real bottom line

183

here by pairing the new fable 5 .2 model with peaceful mcp gemini omni flash

184

and my custom skill we built both a 2d

185

and a full 3d animated short in minutes without touching blender without opening a code terminal

186

and without writing a single technical prompt for a total of just roughly 625 credits

187

now speaking of god a 20 dollar monthly subscription is more than enough

188

but before you go you need to understand one thing nobody explains

189

and it's the real reason i made this video today right now animated shorts

190

and cartoon talking head videos are pulling millions of views across different social media platforms

191

and you don't even have to stick to stickman characters with this exact workflow

192

you can swap in any character sheet you want

193

and create your own signature 2d

194

or 3d cartoon style you can use this to build a faithless youtube channel

195

and monetize through ad revenue you can use it to grow your own personal brand

196

or you can sell these animated videos to other creators

197

and companies who want viral shorts without paying thousands of dollars to an animation studio on top of

198

that big ai companies are constantly sliding into dms of creators

199

who post clean animated stories like this to pay for sponsored collaborations

200

now you have the exact same tools on your laptop right now

201

and you have everything it takes to pull this off too like i promised at the start everything from today's video.

202

The full setup guide, the character issues,

203

the scripts, and my custom Stickman Shorts AI skill is packed in the free guide linked in the description below.

204

Open it up, drop the skill to claw, and test it out today.

205

Tell me in the comments, which version do you like better, the 2D Stickman animation or the 3D version?

206

Smash that like button, subscribe to my creative space, and tag me, Kai Creators,

207

when you post your works because I really want to see what you create.

208

Believe in yourself, go make some art, and I'll see you in the next one.

209

Bye.

下载应用

流利口语所需的一切

AI 发音评分逐句打分

音标练习掌握每个音

词汇学习建立你的词库

词汇游戏边玩边学

扫码

Download on theApp Store Get it onGoogle Play

同频道

How To Make Insane Video Editing With Opus 5.5 (Claude Motion)

Daily ConversationUniversity StudentsWorking Adults

FOR YOU

本课的词汇与口语要点

这节口语课以视频“How to Make Viral Stickman Animation With Fable 5.2”为素材。视频中反复出现的词有：animation, fable, exact, character, stickman。这段视频共有 209 个句子、2657 个单词可供跟读。讲话部分时长为 14:35。说话人语速较快，每分钟约 182 个词，会出现较多连读和弱读。只有 79% 的单词属于英语最常用的 3,000 词，词汇难度较高。

视频中的重点词汇

视频中 15 个值得学习的单词，附发音和释义：

单词	发音	释义
animation 名词	/ˌæn.ɪˈmeɪ.ʃən/	動畫 /动画, 動漫 /动漫
cloud 名词	/ˈklaʊ̯d/	йүн, 雲 /云
sheet 名词	/ʃit/	篇, 張 /张
skill 名词	/skɪl/	技巧, 技能
creator 名词	/kɹiːˈeɪ̯təː/	造物主
custom 名词	/ˈkʌstəm/	習慣 /习惯, 習俗 /习俗
script 名词	/skɹɪpt/	文章
format 名词	/ˈfɔːɹ.mæt/	格式, 開本 /开本
flash 动词	/flæʃ/	閃光 /闪光
edit 动词	/ˈɛdɪt/	編輯 /编辑
silicon 名词	/ˈsɪləˌkɑn/	硅, 矽
setup 名词	/ˈsɛtʌp/	陷害
tutorial 名词	/ˌtjuːˈtɔːɹɪəl/	輔導 /辅导, 教程
bother 动词	/ˈbɑðɚ/	煩擾 /烦扰, 麻煩 /麻烦
connect 动词	/kəˈnɛkt/	連接 /连接

视频中出现的短语动词

单词	释义
turn out 动词	原來 /原来
break down 动词	壞了 /坏了, 出故障
eat up 动词	吃光
find out 动词	找出, 查明
leave behind 动词	遺棄 /遗弃
sign up 动词	報名 /报名, 註冊 /注册
step down 动词	卸任
take over 动词	接管, 接手

视频中的语法

说话人最常用的结构，并附上视频中的原话：

结构	视频中的用法
被动语态 be + 过去分词 — 强调发生了什么，而不是谁做的	is officially left · are actually subscribed · is proven
现在完成时 have/has + 过去分词 — 过去发生但与现在仍有关联的事	haven't heard · you've never opened · has generated

需要注意的发音

说话人用了 35 次缩略和弱读形式，例如 don't, I'm, I'll。请按听到的简短形式来说。

“th” 音: bother /ˈbɑðɚ/, thin /ˈθɪn/, algorithm /ˈælɡəɹɪðm̩/
“sh” 和 “zh” 音: animation /ˌæn.ɪˈmeɪ.ʃən/, sheet /ʃit/, flash /flæʃ/, spatial /ˈspeɪ.ʃəl/, contrast /kənˈt(ʃ)ɹæst/
长单词——注意重音位置: animation /ˌæn.ɪˈmeɪ.ʃən/, incredibly /ɪŋˈkɹɛdɪbli/, collaboration /kəˌlæbəˈɹeɪʃən/, notification /ˌnoʊtɪfɪˈkeɪʃn̩/, iteration /ˌɪt.əˈɹeɪ.ʃən/

中文母语者容易读错的音：

/v/ — 不要读成 /w/，上齿轻触下唇: viral /ˈvaɪɹəl/, actively /ˈæktɪvli/, preserve /pɹəˈzɝv/, reinvent /ɹiːɪnˈvɛnt/
词尾辅音 — 要读清楚，后面不要加元音: cloud /ˈklaʊ̯d/, sheet /ʃit/, generate /ˈd͡ʒɛn.ə.ɹeɪt/, script /skɹɪpt/, format /ˈfɔːɹ.mæt/
/r/ 和 /l/ 的区别: viral /ˈvaɪɹəl/, tutorial /ˌtjuːˈtɔːɹɪəl/, roughly /ˈɹʌf.li/, scroll /skɹoʊl/, carefully /ˈkɛɹf(ə)li/

如何用这段视频练习

先完整听一遍视频，不要开口，记下不认识的单词。
先用 0.75 倍速逐句跟读，熟练之后再回到正常速度。
录下自己的声音并与原声对比，特别注意 animation, cloud, sheet 这类单词。

什么是跟读法？

跟读法 (Shadowing) 是一种有科学依据的语言学习技巧，最初开发用于专业口译员的培训，并由多语言者Alexander Arguelles博士普及。这个方法简单而强大：您在听英语母语原声的同时立即大声重复——就像是一个延迟1-2秒紧跟说话者的影子。与被动听力或语法练习不同，跟读法强迫您的大脑和口腔肌肉同时处理并模仿真实的讲话模式。研究表明它能显着提高发音准确性，语调，节奏，连读，听力理解和口语流利度——使其成为雅思口语备考和真实英语交流最有效的方法之一。

影子跟读法: 阅读完整分步指南 →

跟读练习: How to Make Viral Stickman Animation With Fable 5.2 (Full Breakdown) - 通过视频学习英语口语

选择练习模式

下载应用

同频道

How To Make Insane Video Editing With Opus 5.5 (Claude Motion)

推荐视频

Claude Opus 5.5 Can Now Make Kurzgesagt-Style Animations (Full Tutorial)

Dear Higgsfield, I reverse engineered you in 12 hours

The New Generation of YouTube Influencers

How to Edit VIRAL Talking Head Videos Using ONLY Claude! (New Minimal Style)

9 Insane Claude Mods I Already Can't live Without

How I Learned Procedural Generation

Fastest way to become an AI Engineer in 2026

NEW Claude AI Can Create Faceless Youtube Videos In MINUTES!

How to Create Viral AI Storytelling Shorts

12 Camera Movements for CINEMATIC FOOTAGE - CREATIVE SHOT IDEAS for BETTER B-ROLL - Video Shot Ideas

10 Years Of UX/UI Experience In One Claude Skill

Animation Basics in 14 Minutes (6 Big ideas for beginners)