Практика Shadowing: AI Animation Pipeline: How I Make Cartoons 100% With AI - Изучайте разговорный английский по видео

Выберите режим практики

Каждый режим открывает полноэкранную комнату практики.

166 предложений

Если предложения слишком короткие или длинные, нажмите Edit, чтобы их изменить.

I just created this full cartoon in only 20 minutes with no animation skills, and I only used AI.

So for years, making a cartoon meant mastering drawing,

animation, editing, voice acting, and somehow keeping everything consistent from episode to episode.

Simply put, it was too much work for just one creator.

But today, that's no longer true.

Studios like Netflix, Disney, script art and character design to animation and rendering.

But here's where things get really interesting.

Many of the same capabilities that used to require millions of dollars in budget are now available to literally anybody.

And I'm not exaggerating, I can tell you with certainty

that this is the first time in human history

when one person can control a production pipeline that used to require an entire animation studio.

So, in the next 20 minutes, I'm gonna show you how you can, from scratch, write a captivating script, build characters with distinctive voices,

and create a world around them that looks original.

So by the end of this video, you'll have a real, unique cartoon with a complete story, and you'll do it all in one sitting.

So without further ado, let's get into it.

Now before we even generate a single image, we first need to decide what this world is actually going to look like.

One mistake that I often see people making is jumping straight into the prompting.

Every great animated show out there starts with art direction first.

So before you create a character, you need to define the visual language that underpins the series.

So here's what I recommend: spend some time collecting references.

Now Pinterest is great for this, but you can use Instagram, ArtStation, basically anywhere that you can find artwork that resonates with you.

Don't look for one perfect image, instead build a library of styles,

colors, character that you're looking to create.

But to save you a few hours, I've already put together a huge collection of references that you can use as a starting point.

You can find the entire folder linked in the description below.

So once you have your visual direction locked in, then let's start building.

Now it's time to find the character who's going to carry your entire series.

And honestly, don't think of this as just generating an image.

Think of it as actually casting your lead actor.

You're going to animate this character over and over again, so don't just settle for the first result.

Start by uploading two things.

A few images of your character, or yourself if you want to be the main character, and one of the style references that we picked earlier.

One's going to tell the AI who to create, while the other will tell it how they should look.

Then, you need to describe everything that you want to stay consistent.

Hair, clothes, accessories, eye color, anything that you leave out will become something that the AI has to interpret.

multiple variations, compare them side by side, and do not stop until one immediately feels like the face of your show.

I've actually also included the exact prompt template to use in the description, so all you need to do is replace my character description with yours.

Now once you've found your character, then we can make them consistent from every single angle.

Like I said, now that we have our character, our goal is to make sure they stay consistent throughout the entire series.

what your character looks like from that specific angle.

As soon as you ask for a different expression, a different pose, or change the camera angle, then it's going to start filling in the missing details on its own.

And that's exactly where you're going to start to experience drift.

By now, you already know the rule: whatever you don't define, the AI will hallucinate.

A character sheet is simply where we lock in those decisions once so every future generation can follow the

final adjustments to the design and then generate your character sheet.

Honestly, this is also the perfect time to fix anything that you don't like before you move on.

It's much easier to do this now than after you've already generated 20 different scenes.

So from this point on, your character sheet becomes the foundation for every image and every video that you create.

Instead of redefining your character every time, you're giving the AI a single source of truth that it can reference throughout the entire series.

Now that we've locked in our main character, we can start expanding our world.

Instead of just starting from scratch every time, use your character sheet as a reference.

This is gonna keep every new generation in that exact visual style while letting you create completely new assets.

For example, I can generate a dog that matches the style we've already built.

Later on, I'll use that same workflow to create locations, props, and honestly, everything else that the story might need.

same visual foundation, everything naturally feels like it belongs together.

And you'll notice that as you do this, your asset library is going to grow very quickly.

So that's why I highly recommend organizing everything from the start.

Every time you generate something that you're happy with, whether it's a character, a location, or a prop, make sure to save it to your favorites.

And trust me, this is going to save you a lot of time down the line.

You'll be able to instantly find the the work for you.

By the end of this step, you should have what's a small library of organized characters, environments, and props that are ready to use.

That preparation is what makes the rest of the production process dramatically faster.

Now that we've built our characters, it's time to build the sets that they'll actually be performing in.

So just like your characters, your locations need to stay consistent throughout the entire series.

I mean, think about shows like The Simpsons.

You can instantly recognize recurring locations are part of what makes the world building complete.

The workflow is almost identical to what we did with our characters.

Start by referencing your character sheet so the AI keeps the same visual language.

Then, generate the master version for your location.

Now in my case, I'm creating a cozy living room, but the same exact approach is going to work for literally any environment that your story needs.

Once you've locked in that location, don't Think about all the camera angles that you'll actually use when you're telling your story.

Maybe one shot is going to focus on the sofa.

Maybe another will look towards the window.

Another perhaps is from above.

And the last one is from the floor where your character might be sitting.

Now, every single one of these is going to become a reusable shot

that you can come back to time and time again.

This is where your production pipeline is going to start to pay off.

The AI is already going to know exactly what your character and the environment should look like.

All that you're doing is telling it which character goes to which location.

Now, this is going to give you a huge amount of control over the final result.

Your character is going to stay consistent.

Your location is going to stay consistent.

And every single scene feels like it's happening inside of one world instead of a completely new one every single time.

Of course, if you just need a quick cutaway, you don't always need to build an entire location sheet you can simply reference your character sheet

and describe the environment that you want.

That's perfect for one -off shots or scenes that appear just one time.

But for locations that you're going to revisit throughout the entire series,

spending a few extra minutes up front building proper location and character sheets will help save you hours later on.

At this point, you've built a complete library of reusable assets,

consistent characters, reusable props, a consistent appearance.

The final step is just giving them a consistent voice.

And in the same way that people recognize a character by the way that they look, they'll also recognize them by the way that they sound.

A unique voice makes every conversation feel that much more believable.

For this, we'll head over to the audio tab and open Seed Audio.

You can either choose one of the built -in voices

or upload your own if you want your character to sound just like you.

Now, what I'll usually do Where's the TV remote?

And I'll compare the results until I find the voice that fits my character best.

100

Once you've found the right one, make sure to download it.

101

We're going to use it every single time throughout the series, so it's worth spending those few extra minutes making sure you've got it right.

102

Simply upload every voice you've generated into your assets, give each one a clear name so that you can instantly tell which character belongs where, and from this point on,

103

every major part of your production pipeline is now reusable.

104

Your characters are saved, your locations are saved, your props are saved, and now your voices are too.

105

That means that every time you build a new scene, you're no longer starting from scratch.

106

You're simply pulling the right assets together to tell the next part of your story.

107

usable assets, it's time to finally put them to work.

108

Now for this part, I'm using Claude with a custom skill that I built for this exact workflow.

109

A skill is basically a set of instructions that Claude is going to follow every single time.

110

And mine already contains our script, character sheets, location sheets, and voice references.

111

So Claude doesn't just know what I want, it knows the entire world that we're working in.

112

And yes, it's linked below in the description, so you can use the same exact one.

113

All that I need to do is is write a simple description of what should happen.

114

Now in this case our character is lying on the couch and they're looking for a mysterious remote hidden between the cushions.

115

Once he finally finds it, every button press instantly transports him to a different point in time.

116

That's it.

117

That's the entire prompt that I fed Claude.

118

The skill is going to take one sentence and turn it into a complete production prompt.

119

It's going to plan the camera shots, break the scene into every single line of dialogue exactly where it belongs.

120

Now, by default, it structures everything as a 15 second generation.

121

If you want a shorter scene, just tell Claude, make it five seconds and adjust everything else automatically.

122

Once the prompt is ready, we send it over to Higgs field, we'll attach our references, the character sheets for everyone in the scene, the location we want, and the voice references that we've prepared earlier.

123

But here's the part most people get wrong.

124

Don't just upload your references what it needs to do.

125

You need to tell it explicitly, and that's what the @ key is for.

126

Now, inside the prompt, whenever you mention a character or a location, hit the @ sign and tag the exact reference you want.

127

So instead of writing, show the man sitting on the sofa, you're writing, show image one sitting on the sofa from view one shown in image two.

128

And that same exact rule is gonna apply to your voices.

129

After every single line of dialogue, hit the @ sign saying it.

130

Now this feels like a tiny detail, but in reality, this is exactly what keeps a multi -character conversation from falling apart.

131

Skip the tags and the model is going to start mixing voices halfway through the scene.

132

Now when everything is tagged, then we can finally generate.

133

Another long day.

134

Where is it?

135

This is not a TV remote, whatever.

136

I can't not press it!

137

Intruder!

138

Seize him!

139

Break it!

140

- -I'm sorry.

141

Okay, okay.

142

That really happened.

143

You stay in there.

144

Bedtime.

145

No more.

146

TV.

147

Every new generation now feels like another episode of the same show instead of a completely different project.

148

One thing that I've learned after generating hundreds of scenes is that you almost never need to start over from scratch.

149

Honestly, if a shot isn't perfect, I'll usually make a small adjustment instead.

150

Maybe I'll swap a key visual, rewrite a single action, or I'll regenerate just one scene instead of the entire sequence.

151

That's one of the biggest advantages of building your assets first.

152

Because your characters, locations, voices, etc. are already locked in, every iteration is going to build on the same foundation instead of creating something completely different.

153

Now you'll also notice that some scenes naturally work better than others.

154

So when you're choosing your key visual, try to avoid anything that's unnecessarily difficult for the model to understand.

155

Simple, readable compositions almost always will produce better results.

156

And if a generation has a small mistake at the beginning or at the end, don't be afraid to fix it in the edit.

157

Sometimes a simple trim is all it takes to turn a good generation into a great one.

158

And here's the biggest mindset shift.

159

Animation studios always build their production pipeline first.

160

They create characters, environments, props, voices, and the visual language that underpins the world.

161

Then, every new episode is simply another story that's been built on top of those same assets.

162

That's actually exactly the same thing that we've built here.

163

Once your pipeline is in place, doesn't mean rebuilding your entire world.

164

It actually just means opening Claude, describing the next scene, relaxing, and letting the workflow do the rest.

165

The skill I showed you throughout this video is actually linked below if you wanna try it for yourself.

166

So as always, thank you guys so much for watching and I'll see you in the next one.

Скачать приложение

Всё для свободной речи

ИИ-оценка произношенияОценка каждой фразы

Практика IPAОсвойте каждый звук

СловарьСоберите свой словарный запас

Игра со словамиУчитесь играя

Сканировать QR

Download on theApp Store Get it onGoogle Play

FOR YOU

Цели разговорной практики для этого клипа

Этот видеоролик помогает улучшить навык описания последовательностей действий и объяснения технологических процессов на английском. Вы научитесь ясно структурировать мысли, использовать специфичную лексику и поддерживать логику в речи — это ключевой этап для достижения свободы в разговорах на темы творчества и инноваций.

Банк фраз

"Simply put" — Проще говоря
"from scratch" — С нуля
"lock in" — Зафиксировать
"side by side" — Рядом друг с другом
"single source of truth" — Единственный источник истины (основная референсная точка)

Эти выражения часто используются в повседневной и профессиональной речи, их освоение поможет сделать ваши разговоры более естественными.

Исправляем слабые места

Особое внимание в видео уделяется ритму и интонации при объяснении сложных шагов. Многие из нас боятся перегружать речь терминами, но в этом диалоге показано, как плавно встраивать их, сохраняя понятность. Практикуйте "shadowing" (повторение за говорящим) — это отличный способ улучшить произношение английского и чувство ритма. Используйте для этого специальные сайты или приложения, которые позволяют повторять фразы в реальном времени. Помните: регулярная практика разговорного английского с YouTube клипами — это эффективный способ прогрессить. Не бойтесь экспериментировать с интонацией: подчеркивайте важные слова, чтобы ваша речь была динамичной и привлекательной.

Учаить английский с YouTube становится проще, если вы фокусируетесь на конкретных навыках. Этот ролик — отличная возможность потренироваться в структурированной речи и расширить словарный запас. Начните прямо сейчас: выберите несколько фраз, повторите их "shadowing" и зафиксируйте результат — вы увидите прогресс уже через неделю!

Грамматика в этом видео

Конструкции, которые говорящий использует чаще всего, с точными словами из видео:

Конструкция	В видео
«Used to» used to + глагол — прошлая привычка или состояние, которых больше нет	used to require
Present Perfect have/has + причастие прошедшего времени — прошлое действие, важное сейчас	I've already put · I've actually also included · you've found
Пассивный залог be + причастие прошедшего времени — важно, что происходит, а не кто это делает	are saved · is tagged · are already locked

Что такое техника Shadowing?

Shadowing — это научно обоснованная техника изучения языка, изначально разработанная для подготовки профессиональных переводчиков и популяризированная полиглотом доктором Александром Аргуэльесом. Метод прост, но эффективен: вы слушаете аудио на английском от носителей языка и немедленно повторяете вслух — как тень, следующая за говорящим с задержкой в 1–2 секунды. В отличие от пассивного прослушивания или грамматических упражнений, Shadowing заставляет мозг и мышцы рта одновременно обрабатывать и воспроизводить реальные речевые паттерны. Исследования показывают, что это значительно улучшает точность произношения, интонацию, ритм, связную речь, понимание на слух и беглость речи — что делает его одним из самых эффективных методов для подготовки к IELTS Speaking и реального общения на английском.

Техника шедоуинга: читать полное пошаговое руководство →

Практика Shadowing: AI Animation Pipeline: How I Make Cartoons 100% With AI - Изучайте разговорный английский по видео

Выберите режим практики

Скачать приложение

Рекомендуемое видео

Is AI About to Change Movies Forever?

100 Days of Bad Drawings

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

How to Create a Character with Samuel L. Jackson | Discover MasterClass | MasterClass

How to Build AI ChatBot with Custom Knowledge Base in 10 mins

NEW Claude AI Can Create Faceless Youtube Videos In MINUTES!

AI Coding Is a Dead End

Creating Iron Man suit. Full workflow by @AlexLab. Real life Iron Man exosuit with a box of scraps)

I Turned MODERN SpongeBob Into SEASON 1…

Dear Higgsfield, I reverse engineered you in 12 hours

How to Create Viral AI Storytelling Shorts

making architecture art

Цели разговорной практики для этого клипа

Банк фраз

Исправляем слабые места

Грамматика в этом видео

Что такое техника Shadowing?