Luyện nói tiếng Anh bằng Shadowing qua video: Dear Higgsfield, I reverse engineered you in 12 hours

Hi, welcome to this masterclass.

My name is Sirio and this is entirely AI.

I'll show you.

So that's me in a scene that I was never actually in and I didn't even use any app to make it.

And this is what actually happens when you try to do it the normal way.

I'm uploading that same template video into cdance 2 .5 and asking it to swap me in.

Blocked.

But apps like Hicksfield's Ganjutsu do this exact thing all day every day with these exact videos.

So how the are they getting around it?

Hey friend, welcome back.

Today We're going to reverse engineer Higgsfield's Ganjutsu

and we're going to get around the IP blocks in C -Dense 2 .5

So that you guys can put yourself into pretty much any movie scene

or talking head you want without any special software A quick heads up before we start, this video is for educational purposes only.

I basically want you to understand how these AI models actually work and what they can really do.

And this might end up being the first episode of a

new series where I actually show you guys the stuff that they do not want you to know.

And by they, you know who I'm talking about.

Okay, so no fluff, here's how this is going to work.

We're going to start by doing the whole thing by hand, okay? it's actually going to break on us twice

and we're gonna fix both times at every step along the way I'm going to pause

and tell you why it works and not just what to click

and then I'm going to give you the AI for the automatic version the agent version though the workflow

so that you can turn it into a template for your own stuff

and in the next videos we're going a lot further

and we're gonna make things like this I'm Kim Kardashian

and this is master class I'm Serio and this is your time to build something of your own cut good job

So what do you think was it any good man you killed it, but first the basics

So let's not waste any more time you guys have all seen these videos

that are basically everywhere right now And the way

that it works is that you upload a photo of yourself

And then you pick a template video

and then it just swaps you in for the person that is in that template video So it's the same moves, the same everything, the same lip syncing.

And most people get this from one of those wrapper apps that sells it as a feature.

And like I've said before, that's really the whole reason that a wrapper works and good for them.

What I mean by that is that they are selling to you an API workflow.

It is not the AI model.

It's the workflow.

It's the thinking.

They are selling you the thinking that they do for you, and you just press a button and so this tutorial is for those people who actually want to build that button.

I spent the last 12 to 24 hours on this going through open source models, masking, cutting people out, and a whole bunch of other subs

that didn't work and I ended up with the one that I'm going to show you today.

So let's go back to that blog generation from the start.

This is the template

and all I did here was I head over to cdance 2 .5 I uploaded this exact video

and I typed replace the character in the video with the picture

that I uploaded um and it got blocked for IP which

which I mean sure it makes sense these people are famous

but like I said this exact video works totally fine in genjutsu

but before we actually build anything I think

that is important for us to understand why because

if you get this part the rest of the video is

basically just just details there are really only two things that get you blocked here the The first one is the face, the second one is the music, and the way that I like to think about this is

that a video is really carrying these two things at the same time.

It's carrying who's in it, so the face, the voice, the music, and it's carrying how things move, so the motion, the timing, and the lips.

And the filter, the IP filter only cares about the who.

It does not care about the how, and the how is the movement.

So the whole idea here is that you keep the how, the movement, and you throw away the who,

the IP, And everything that we're going to do from here on out is really just hide one of those two things, obviously, without losing the movement.

So the first thing

that we're going to have to do is turn this template into something where the AI cannot tell who the people are, but he can still see how they move.

And for that, we're going to use something that is called Depth Anything.

You can find it on Hugging Face or Replicate or Foul.

And all you have to do is that you upload your video, you click Generate, and then you get something like this.

So this is what's called a depth mask, and what I mean by this is that it throws away what everything looks like,

and it keeps how close or how far things are from the camera.

Like, it's the depth.

If you've used Photoshop or you've done retouching before, it's kind of very similar to Dodge and Burn, but...

In this case it tracks all the movements the movements are still there, but the faces as you can see They're not just a silhouette this one is from foul

and this one is from replicate I like the one from replicators.

I don't know.

I just like the colors of it

There's no difference you can use the one from foul too

and there's one more thing Which is the audio right if we want the lips to match the song?

Then the audio has to go through as well.

Now the first thing that most people would say is just mute it

But we cannot do that and the reason why is

that AI actually needs to hear the voice in order for it to move the mouth So instead of muting it, we're gonna keep the voice

but get rid of everything else Keep the actual words remove the soundtrack remove the background music

So what I'm going to do is

that I'm going to take this depth mask and bring it into CapCut

and I'm just going to make a new project

and drag the clip in and then I'm going to click on it

and over here on the right hand side I'm going to

go to audio then voice changer then voice filters then I'm going to pick pitch up.

This is a voice filter

and what it does is it just changes the pitch of the music

and then once that's done I'm going back to the basics.

I'm gonna click on isolate voice Then keep only the vocals.

And so now what's left is basically only the acapella without the music or soundtrack So listen to it.

Let's get it Hop off a 16 passing the D5.

No, that's not a challenger.

I get some members with me in the fridge get cold seat that's some cannibals

so that's both of them hidden right cdense 2 .5 isn't going to block this anymore

because there are no faces there's no music

and we still have all of the movement right pretty neat

now instead of giving it the original video we're going to

give it the depth mask video plus our own photos

and then swap the people in but there's one problem though

100

which is that the depth mask has no idea what the place actually looks like.

101

It's just either like a grayscale version of the depth mask like in FAL or like this other one here.

102

And I think it's easier to show you guys.

103

So let's run a quick test.

104

In order to do that, we need an API key.

105

So I'm going to grab my enhancer API key for Cdance 2 .5 so we can test.

106

And the link for that is in the description.

107

And yes, it is cheaper than all of the other platforms that say that they are the cheapest.

108

and also it doesn't train on your data

109

so your stuff is safe i'm going to copy the docs right here

110

and then i'm going to paste them into codex

111

or claude i'm using codex right now that's really

112

because i already had this project started in here it does not mean

113

that i'm endorsing openai

114

or anything it's just easier for me right now hey can

115

you set this up for me then i'm going to head a key copy it

116

and I'm gonna go back to the agent

117

and I'm going to ask it to make an ENV file

118

which is just like a little private file in your folder

119

that holds your keys and and saves the keys in there for safety

120

so please guys never ever paste your keys in the chat

121

always put them in an ENV file I'm gonna click on here

122

and I'm gonna paste in my API key and we're all set and we are ready to generate.

123

But before we actually make anything, the next thing I want to talk about is draft mode because a lot of people get this wrong.

124

What I mean by draft mode is C -Dense 2 .5

125

lets you make a 480p version of your video first over

126

and over until you like what you see and then it finishes

127

that exact same video in 1080p so you don't have to lose any detail.

128

And it's not like an upscaler, It's kind of like doing a pencil sketch before you paint.

129

You can redo the sketch over and over again as many times as you want And then when you're ready, you're like cool I'm gonna pay for the paint once

130

so use draft mode every time that you're trying to draft something

131

Or generate something with C dense now Let's say

132

that you get a draft

133

that you really like what a lot of people will do is they're gonna download it

134

and then upload it right back as

135

The reference video thinking that's the same thing it is not

136

and the reason why is that the draft is tied to a task ID

137

and not the video file so when you upload

138

that file again that's actually totally different kind of request

139

and it doesn't carry over any of your settings

140

so there's no promises that you're gonna get the same video back

141

so basically you do all of the cheap tries in draft mode

142

and then you only pay for 1080p one time hopefully this makes sense you're gonna see it in action

143

so I'm just going to drag in the depth mask this one

144

and then the two photos that I want to use

145

which is me and teen and

146

if you guys haven't met teen yet teen is this a little creature

147

that I've made with AI basically my companion and also a really cool new project that I'm working on

148

that is also linked in the description as well

149

and then i'm going to paste in this really simple replacement prompt you can screenshot that

150

and then i'm just basically going to say hey use the

151

draft mode for this generation um cdens also needs links not files

152

so please throw everything on a temporary server like cat box

153

or temp files it's going to host your inputs temporarily

154

and then you're going to hit enter

155

and this usually takes about one minute okay cool

156

so here's what we got and

157

if you want to grab the video you can either get it from the agent

158

or you can go into the logs and usage in your enhancer dashboard

159

and give it a quick refresh

160

and it will be sitting right there with your request id that's also going to match your agents

161

and i mean it's it's not great right well we basically have two problems here the first one is

162

that it didn't actually swap me in all the way.

163

So that's my face, but that it's very much the actor's body with all the long hair.

164

And the second thing is that the whole thing is just gray.

165

So one way to fix this is to have the agent describe the studio.

166

And the reason why is, like I said earlier, the depth mask has no idea what the place looks like

167

and if this was your reference video this is what you got back

168

so that's where the gray comes from

169

and what i'm going to do is i'm going to take a screenshot of the original video

170

and i'm going to drop it into the agent

171

and i'm going to say hey describe this place for me

172

but do not describe the people and then update the prompt with

173

that and send the same request again so that's the problem

174

that i'm using you can pause

175

and screenshot this since we're doing this with the agent the agent knows exactly what they sent before

176

and they can just update that request or update

177

that payload you don't have to do anything manually

178

because the agent has context of your workflow

179

and i'm just going to hit enter and submit it okay now we have color

180

which is nice um that's one of our two problems fixed

181

but i mean the other problem is still there For some reason, it is still sticking way too close to the depth mask,

182

which is why it's still the actor's body.

183

It's not mine.

184

I'm not entirely sure why.

185

I think it's basically, I don't know, I think it just sees an outline and then it traces it.

186

But I've got one more trick to fix this.

187

So this time, I'm going to grab the original video and then our depth mask with the changed audio.

188

I'm going to drop the original clip into cap cut

189

and then I'm going to put in the depth mask right

190

on top of it I'm going to mute the original clip

191

So the only audio left is our pitched vocals

192

and I'm going to Click on the depth mask

193

and go over to the video tab

194

and the background removal Then click on custom removal

195

and just highlight the people Basically what this is doing is

196

that it's cutting out just the people

197

and then it's laying them on top of the original video now why does

198

that work well i well

199

if you think about it like now it's it's a very different video we were giving it a depth mask

200

which is really just the outline so it was tracing the outline before

201

and now we are giving it something that that is totally different

202

and it stops from tracing it

203

so i don't know it looks at the whole thing again

204

and i'm gonna export

205

that this is what it looks like then i'm gonna send it back to the agent

206

and say hey swap the old source video for this new one

207

and make another draft and look at that

208

that worked that's pretty cool okay so

209

that was basically four moves we hit the face we hit the music we describe the room

210

and then we laid the people over the real video

211

and that's way too much manual work to do every single time

212

but but you can do all of it with one click the same way Hicksfield does with a workflow

213

so I had the agent turn everything we just did into one all you need is a replicate account

214

and I'm giving you a skill

215

that sets everything up hugging face is also optional just a backup choice

216

so really quickly it is 11 steps one we upload the

217

original video we keep the original video for also the final audio.

218

Two, Demicus on Replicate splits the vocals from the music.

219

Three, rubber band pitches the vocals up by three semitones.

220

Four, we have Depth Anything video, which makes the colored depth video that we talked about.

221

Five, we have Sam 3 cutting out all the people on every frame.

222

Six, it puts those depth people over the original background.

223

Seven, the pitched vocals go right inside of

224

that video eight you check everything before it sends nine see

225

dance makes a draft three enhancer in the edit mode ten it throws away see dance audio

226

and puts in the whole original soundtrack back

227

and eleven you approve of it and it makes a 1080p video and you download the final mp4 I know,

228

well, but this workflow is making millions of dollars.

229

Now you don't have to do any of that yourself.

230

I'll show you all of it running as soon as it is set up.

231

So this whole thing runs through your agent and it comes with its own interface.

232

I built all of it for you guys.

233

All you have to do is download the skill, give it to your agent, and then plug in your API keys from Enhancer and Replicate so you can grab the skill in the description, grab the folder, and we're gonna drag it in,

234

in codex, and we're gonna say, please read the readme file and set this up for me

235

and just let the agent do it the agent is gonna set up everything for you

236

and ask you some questions along the way and

237

if you don't want to download the files for whatever reason

238

you can just ask your agent to build this workflow over here

239

so you can screenshot this

240

or just give it my video transcript of this YouTube video now I already have mine set up

241

and there are two ways to run this the first one is you just drop in the video

242

that you want to change and then the photos

243

that you want to be placed in the video or like swapped and you say

244

hey run the workflow as simple as that the second one you do it yourself in the ui interface

245

and that's what i'm going to do here

246

so you guys can see every step

247

so the interface is really just four steps

248

because you don't see everything that's happening in the back end you have to upload review draft

249

and export or save so i'm going to upload my video

250

which can be an mp4

251

or a MOV up to 30 seconds I'm going to leave it on advanced

252

because that's the depth workflow we just built

253

and then I'm going to click prepare and

254

so right now it's doing steps through one to seven all

255

by itself you don't have to do anything it's splitting the

256

vocals from the music it's pitching them up everything we did in CapCut

257

and then it's making the depth video cutting the people with SAM3 laying them over the original background

258

and cleaning up every little tray or edges that are off from

259

that mask this is step eight

260

and it's not going to let me generate until i've watched the whole thing

261

and i've approved it myself and the reason for that is

262

that might be a bad mask um

263

so we don't i don't want you to uh basically just

264

throw money down the drain ai agents are unpredictable that's why we make sure

265

that they check their work every time So it's checking for holes in the body, the hair or hands.

266

This one to me looks very clean So I'm gonna prove this and then I'm gonna add my characters.

267

So character one

268

That's me

269

Character two is optional in this case it's teen

270

and then the prompt is already written And you can edit that if you wish and I'm gonna generate my draft.

271

Cool.

272

So here's our draft and listen to that

273

10 principles that build billion dollar brands take notes that's the original audio now the pitched one um

274

and i'm pretty happy with this so i'm going to approve

275

that and then it makes the 1080p

276

and puts the original audio back one more time um and

277

that one does cost extra and it's never going to do it without asking you first

278

or approving now that workflow is the best one

279

if want to bulletproof your generations

280

and it mostly uses the audio to match the words so

281

if your video has no audio it will still generate for sure

282

but there's one more workflow for

283

when you want the face to move exactly like the original

284

and it's the same workflow but we add what's called a face mash

285

and what i mean by that it's a little wireframe

286

that gets drawn right on top of the face it's kind of like those dots

287

that they put on actor spaces in the behind the scene videos

288

so can track them so the dots can capture how the face moves

289

and then the computer can um totally create a different character on top of

290

that so that same thing same thing applies here um it's going to track how the face moves

291

and it's still going to hide the person behind it

292

and this is actually the easiest one because it only has two steps

293

so you can change the pitch you can add the face mash

294

and there's no cutting anyone out like

295

that like it just removes all those extra other step so

296

if you guys downloaded the project you can just ask the agent to use this workflow instead it's faster

297

and it's also right here in the interface it's called fast

298

or you can ask it to install the face mask tracking

299

and put it on top of the original video you send

300

that in and you get something like this that's dope that's crazy

301

so what the face match is doing is it's tracking 478 points on the face

302

so around the eyes the lips the nose the whole face

303

and it is drawing that mesh on every single frame

304

and it saves all of those points with all the timing in a file

305

that we have set up to be called face landmarks dot json everything is done

306

and saved for you you don't have to worry about it and it uses google's face landmark model

307

from google's official model storage One heads up though, at least right now, this works best with one person, which is centered, talking to the camera.

308

If you have two people moving around, I would just stick with the depth mask workflow.

309

And if you want to bulletproof these even more, you can ask the agent to put the face mask on top of the cutout depth video.

310

So on top of this one right here.

311

And then you get something like this.

312

And then you send that in as a source video. And again, all of this works by just talking to it there's nothing

313

for you to install yourself you just ask the agent to set up the Google MediaPipe face landmarker

314

or any other packages okay so this has been a long video

315

so from here you guys can keep making things better you can add a transcriber

316

so it writes down what's said in the video and puts

317

that into the prompt you could have the agent watch the original video

318

and describe the movement in detail second by second

319

and then you can even use your own voice

320

So you basically make your own avatar look at this

321

and I'm here to spill the secrets behind api's So that you can build million -dollar workflows.

322

That is it And if you take one thing away from this video is

323

that everything every one of these tricks was basically just the same idea Implemented in different ways

324

and methods you just keep how things move

325

and you throw away who is in the video

326

And hopefully you guys know how Genjutsu actually works now

327

and how to build it yourself I have attached the entire open source project in the link in the description

328

So just get it from there.

329

It's free and go make something with it and in the next one We're going to go way past the basics.

330

See you guys there and do not forget create without limits.

331

This is Syria

✨ Video gợi ý

409 câu

Từ vựng và ghi chú luyện nói cho bài học này

Bài luyện nói này dựa trên video “Dear Higgsfield, I reverse engineered you in 12 hours”. Video này có 197 câu và 4127 từ để luyện shadowing. Phần lời nói dài 21:43. Người nói nói nhanh, khoảng 190 từ mỗi phút, nên sẽ có nhiều chỗ nối âm và nuốt âm. 87% số từ nằm trong 3.000 từ tiếng Anh thông dụng nhất, nên từ vựng dễ theo.

Từ vựng quan trọng trong video

15 từ đáng học trong video, kèm phiên âm và nghĩa:

Từ	Phiên âm	Nghĩa
mask danh từ	/mæsk/	mặt nạ
audio tính từ	/ˈɔ.di.oʊ/	âm thanh, âm tần
upload động từ	/ˈʌpˌloʊd/	tải lên
generate động từ	/ˈd͡ʒɛn.ə.ɹeɪt/	sinh ra, gây ra
vocal tính từ	/ˈvoʊ̯.kəl/	nói to tiếng
swap động từ	/ˈswɑp/	giao hoán
template danh từ	/ˈtɛm.plɪt/	bản mẫu, khuôn mẫu
filter danh từ	/ˈfɪltɚ/	máy lọc
approve động từ	/əˈpɹuːv/	phê duyệt
dense tính từ	/ˈdɛns/	đặc, dày đặc
pasta danh từ	/ˈpɑ.stə/	mì ống
screenshot danh từ	/ˈskriːn.ʃɑt/	ảnh chụp màn hình, hình chụp màn hình
dragon danh từ	/ˈdɹæɡən/	rồng
clip danh từ	/klɪp/	kẹp (cái), ghim (cái)
lip danh từ	/lɪp/	môi

Cụm động từ bạn sẽ nghe

Từ	Nghĩa
throw away động từ	vất, vứt
clean up động từ	quét dọn, quét tước
write down động từ	viết ra, ghi xuống

Những câu đáng luyện lại

Các câu ngắn, trọn ý trong video mà bạn có thể dùng lại khi giao tiếp hằng ngày:

I'm going to copy the docs right here.
I'm using codex right now.
Hey, can you set this up for me?
You're going to see it in action.
That's also going to match your agents.

Ngữ pháp trong video

Những cấu trúc người nói dùng nhiều nhất, kèm đúng cụm từ trong video:

Cấu trúc	Trong video
Thì hiện tại hoàn thành have/has + quá khứ phân từ — việc đã xảy ra nhưng còn liên quan đến hiện tại	I've said · you've done · haven't met
Câu bị động be + quá khứ phân từ — nhấn vào việc xảy ra, không phải người làm	is called · is also linked · is set
Mệnh đề quan hệ who / which + mệnh đề — thêm thông tin về người hoặc vật	people who actually · thing Which is · video, which makes

Phát âm cần chú ý

Người nói dùng 108 dạng rút gọn, ví dụ I'm, we're, don't. Hãy nói theo dạng ngắn đúng như bạn nghe.

Âm “sh” và “zh”: mash /mæʃ/, screenshot /ˈskriːn.ʃɑt/, optional /ˈɑp.ʃə.nəl/, capture /ˈkæp.(t)ʃɚ/, refresh /ɹɪˈfɹɛʃ/
Từ dài — đặt trọng âm cho đúng: temporary /ˈtɛmp(əˌɹ)ɛɹi/, automatic /ˌɔː.təˈmæt.ɪk/, temporarily /ˌtɛmpəˈɹɛɹəli/, manually /ˈmænj(u)əliː/

Những âm người Việt thường gặp khó:

Cụm phụ âm cuối — đọc đủ từng âm, đừng nuốt âm cuối: mask /mæsk/, dense /ˈdɛns/, prompt /pɹɑmpt/, export /ˈɛks.pɔɹt/, replacement /ɹɪˈpleɪsmənt/
Âm /tʃ/, /dʒ/ và /ʒ/ — tiếng Việt không có, đừng đọc thành “ch” hay “gi”: pitch /pɪt͡ʃ/, generate /ˈd͡ʒɛn.ə.ɹeɪt/, sketch /ˈskɛt͡ʃ/, package /ˈpæk.ɪd͡ʒ/, generic /dʒɪˈnɛɹ.ɪk/
Âm /æ/ — mở miệng rộng hơn âm “e”: mask /mæsk/, mash /mæʃ/, dragon /ˈdɹæɡən/, soundtrack /ˈsaʊndˌtɹæk/, landmark /ˈlændmɑːk/

Cách luyện với video này

Nghe hết video một lần, chưa cần nói, và ghi lại những từ bạn chưa biết.
Bắt đầu ở tốc độ 0,75×, nói đuổi từng câu, rồi quay lại tốc độ bình thường khi đã quen.
Ghi âm giọng mình rồi so với bản gốc, chú ý các từ như mask, audio, upload.

Phương Pháp Shadowing Là Gì?

Shadowing là kỹ thuật học ngôn ngữ có cơ sở khoa học, ban đầu được phát triển cho chương trình đào tạo phiên dịch viên chuyên nghiệp và được phổ biến rộng rãi bởi nhà đa ngôn ngữ học Dr. Alexander Arguelles. Nguyên lý cốt lõi đơn giản nhưng cực kỳ hiệu quả: bạn nghe tiếng Anh của người bản xứ và lặp lại to ngay lập tức — như một "cái bóng" (shadow) đuổi theo người nói với độ trễ chỉ 1–2 giây. Khác với luyện ngữ pháp hay học từ vựng bị động, Shadowing buộc não bộ và cơ miệng phải đồng thời xử lý và tái tạo ngôn ngữ thực tế. Các nghiên cứu khoa học xác nhận phương pháp này cải thiện đáng kể phát âm, ngữ điệu, nhịp điệu, nối âm, kỹ năng nghe và độ lưu loát khi nói — đặc biệt hiệu quả cho người luyện IELTS Speaking và muốn giao tiếp tiếng Anh tự nhiên như người bản ngữ.

Phương pháp shadowing: đọc hướng dẫn từng bước đầy đủ →

Luyện nói tiếng Anh bằng Shadowing qua video: Dear Higgsfield, I reverse engineered you in 12 hours

I think AI slop is about to trigger a renaissance, actually

DETROIT Become Human - Deviant Interrogation Scene - Best Outcome

OpenAI’s lead engineer was secretly a spy...

DeepSeek R1 Explained in 2 Minutes

Chapter 3

I Tested Secret Technology

Is Siri AI Usable for English Learning?

AI Can Write Code. And That’s the Problem.

AI Animation Pipeline: How I Make Cartoons 100% With AI

hellofriend.mov | Mr. Robot

Therapy for the Vibe-Coded Brain

Philomena Cunk vs Prof. Jim Al Khalili CBE (2025 Edition)