Shadowing-Übung: 11 AI Breakthroughs You Didn’t Know Existed - Englisch Sprechen Lernen mit Video

Übungsmodus wählen

Jeder Modus öffnet einen Übungsraum im Vollbild.

385 Sätze

Wenn Sätze zu kurz oder zu lang sind, klicke auf Edit, um sie anzupassen.

So every week I like to make an end -of -week AI news breakdown video where I share everything

that happened in the world of AI for the week.

Usually those videos are very focused on some of the bigger labs.

They're the ones that get most of the news.

OpenAI, Anthropic, Microsoft, Google, NVIDIA, companies like that.

However, there's often some really cool news that flies under the radar.

There's a lot of research going on and smaller tools that not as many people, including myself, are talking about right now.

So in this video, I want to share some of the cooler things

that I've come across recently that haven't made the bigger end -of -week news videos.

Let me know in the comments if you like this style of video,

and I'll probably make them more often and share some stuff that's not from the big labs.

But let's start here with Fish Audio.

Fish Audio released a model called Drama 3, and it's a text -to -speech model where you direct the

performance in plain English instead of picking from like a menu of emotional presets.

So like most voice tools, they'll give you a dropdown for happy, sad, angry, things like that.

Drama three, however, it lets you write the direction the way you'd like explain it to an actor.

You put it in square brackets right in the script.

So for example, here's some spoken text with no direction.

I'm fine.

Really go on without me.

And now here's one where there's some direction written in, barely holding it together, forcing a smile.

I'm fine.

Really.

Go on without me.

And you can even pick specific words.

Like right here, they specifically tell it to whisper, go on without me.

I'm fine.

Really.

Go on without me.

So you can like scope and affect two specific words.

You could drop in sound effects, things like a small laugh, shaky breath, and you could put them exactly where you want them.

I'm fine.

Really, go on without me.

You can even have multiple speakers with multiple voices.

You were there that night.

I was home, asleep.

Then why is your coat still wet?

It rained.

They even licensed

Uncle Roger's voice for the demo page, which is kind of hilarious.

Haiya!

Nephew, why you put ketchup in egg fried rice?

Fried rice need soy sauce, need MSG, need wok hay, not tomato sauce uncle roger cannot watch this

if you've ever watched like any of the uncle roger videos it's actually pretty spot on

if we take a quick peek at artificial analysis here we can actually see

that fish audio drama 3 here scores the highest on pronunciation

so it's really good at pronouncing words even scoring higher than 11 labs

and spacex ai's text -to -speech model which is one that i actually demoed in last week's news video.

However, when looking at the arena leaderboard preference,

which, you know, sort of did a blind taste test where people listen to a bunch of different models

and pick their favorites without knowing what model they were picking.

Well, it didn't quite do as well.

So, you know, it reads hard words better than almost anything else, but a lot of people still prefer other voices.

It also appears to be open weight available on hugging face.

It is a 36 billion parameter model.

So you're going to need some pretty decent hardware in order to run it.

People have reported that you need roughly a 46 gigabyte VRAM card,

or you can get away with 32 gigabytes with a community quantitized four bit version,

but neither of those are going to run on like most laptops.

A couple other quick notes that I found about this one.

So the license is free for commercial use, but only if you're making under half a million a year

which is probably a non -issue for most people the api is about 10 per million characters right now

but that is a launch discount that jumps to 50 after november 12th

and fish audio takes do vary so plan on generating a few

and keeping the best one and

if you are going to use the api double check

that you set the model to drama 3

because it often will quietly fall back to older models

and you'll still get billed even if you're using the older models

so make sure you check that but

if you head on over to fish .audio slash drama 3 click try drama 3 for free

you can play around with it right now and

if you make audiobooks or game dialogue

or anything with characters probably worth trying out all right

while we're talking about voices once you have

that voice performance you might want like a face to go with it well ByteDance recently released OmniHuman 2 .0.

Now this model will take a single image and an audio track,

and then it'll generate a video of the person talking or singing or acting it out.

Now this is the 2 .0 version.

And the biggest difference is that the previous version topped out at just 60 seconds.

This one will allow you to go up to five minutes in one continuous shot.

So you don't have to like stitch a bunch of short clips together.

and hope that you don't see the seams or hear the seams.

So here's some examples.

The kingdoms called it a legend.

Tonight, I make it real.

And it even works with singing. Did you catch that?

You can also add in text prompts to control camera moves or add things that weren't in the original photo.

Now it generates in 720p resolution and it doesn't just have to be humans.

It can work with like cartoon characters and animals and things like that as well.

Now, if I'm being totally honest, I still think these lip syncing models have a long way to go.

When I watch a video with like an AI lip sync attached to it, it still just like isn't working for me.

We're getting closer and closer.

I mean, like we're way better than we were a couple of years ago, but it still has a lot to be desired.

I don't know.

What do you think? let me know in the comments.

As far as pricing goes, the price on this one dropped from 12 cents a second

in the last model down to seven cents per second in this model.

So if you are making a full five minute video, it's still going to cost you about 21 bucks.

100

Now, a few other quick things to note about this one.

101

Everything I'm showing you is ByteDance's own demo reel.

102

And some early viewers did notice like background shifting around in their clip.

103

Again, it's only 720p you're not going to get 1080p or 4K out of this.

104

And because this is only one photo plus any audio, there's some pretty obvious potential deep fake concerns with stuff like this.

105

Although if you saw what I just saw, I don't think we have to worry about too many people being fooled just yet.

106

This one is available now through Byte Plus and through their API.

107

All right, now let's talk about one of my more favorite topics recently.

108

I've been absolutely loving these.

109

Let's talk about world models.

110

So these are AI models that generate a video world that you could move around in.

111

And Tencent's Arc Lab released one called World Crafter recently.

112

And this one starts from just a single image or a single line of text.

113

Now, one of the big issues that we've had with world models is

114

that if you're inside a world model and you look away from something and then turn back, a lot of times it's like gone

115

or changed the model only remembers the last few seconds of what it showed you

116

and if you move around the room everything kind of goes wonky now world crafters fix is a memory

117

that you look up by camera position so

118

when you point the camera somewhere it pulls the past views

119

that match where you're looking and feeds those back to the generator

120

so when you turn back to the original restaurant it remembers what it saw the first time

121

so they prove it with this little trick here they take the video the model generated

122

and rebuild a 3d point cloud from it if the world were inconsistent

123

that 3d reconstruction would come out as just like mush instead you get clean geometry

124

and the camera path lines back up

125

when it revisits a spot it handles moving things too like a character going out of frame

126

and then later actually coming back into the frame like we

127

can see with this horse video here where the camera takes the horse out of frame

128

and then when the camera pivots back around to the right you can see

129

that the horse is still there even though it was moving

130

now on their own benchmark of 145 different scenes they say

131

that it beats the other models they tested on revisit consistency

132

and camera control but just keep in mind that that is their benchmark so you know take that with a grain assault.

133

The code for this is available on GitHub right now,

134

and there's even a version on hugging face that you can go and play around with right now.

135

In my own testing, though, telling it what to generate, it's kind of not the fastest thing in the world, but eventually it gets there.

136

ByteDance also showed off their own version of a world model recently at an event in Beijing.

137

My buddy, Tim from Theoretically Media, he posted some footage that somebody sort of took during the event.

138

So it's like this bootleg camera version.

139

Now, the wildest demo they shared was they took a scene from Stephen Chow's movie,

140

King of Comedy, and generated a brand new camera path through it.

141

It's like reshooting a shot that was filmed decades ago.

142

You can now get new angles on that shot.

143

You can also type a description to get a world,

144

spawn a giant spoon, and then climb it and walk through a portal from a room into outer space.

145

Now there is a catch and that's that we can't use it yet.

146

It doesn't even have a public name yet.

147

And there's no release date.

148

And the only footage I've actually seen of it was from people who recorded it at this event as you know,

149

Tim from theoretically media shared here.

150

So yeah, we'll keep an eye on that one, but I'll see you soon it looks pretty impressive real quick before

151

we get back to the video

152

i wanted to show you guys how i've been using

153

this new feature dots from openai now i've talked about this in a recent ai news video

154

and since then i've dug into it even more basically dots are these always on ai agents

155

that you can assign ongoing responsibilities to

156

and they keep on working on them even

157

when you're not using chat gpt anymore

158

so for example i've been using my dot to help me with my youtube analytics

159

and updating them automatically every time I post.

160

I connected it to my YouTube channel and it pulls views

161

and engagement data to find patterns across my titles and topics and angles and post times.

162

And I also have a shared space with my team where my dot keeps all this organized

163

and creates new sub pages with weekly reports and video ideas based on what's working.

164

What's cool is that each dot has its own cloud computer

165

and can use chat GPT work

166

and codecs to get things done it's powered by gpt6 astra

167

and learns my preferences so it can proactively suggest things without me having to ask

168

so check it out in the link in the description

169

and thanks so much to open ai for sponsoring this portion of today's video all right

170

so we already saw two world models trying to fix the

171

memory problem well skyworks matrix game 4 .0 takes a completely different approach

172

this one stops asking the ai to remember the world at

173

all with a normal world model the ai is doing everything it's drawing the frame

174

but it's also the physics engine the game rules and the memory

175

when something leaves the screen the only record it ever existed is in the pixels the model already drew

176

and that's kind of why things tend to vanish matrix game splits

177

that job into two an ordinary game engine runs the game it knows where every enemy is what happens

178

when you shoot something what's behind you it draws a cheap blocky low poly version of

179

that scene then the video model's only job is to repaint

180

that blocky frame so it looks finished

181

so we can see in some of these models down here

182

on the right is what the game engine actually draws on

183

the left is what you see it's the same frame in the same moment

184

but the engine handles the gameplay and the AI handles the look that's skinned over the gameplay.

185

Since this game engine is keeping track of everything, the world's not going to forget a pyramid behind you is still in the engine's map,

186

whether the AI is drawing it back in or not.

187

Here's another example of a car game.

188

You can see the blocky early rendered version and then AI putting its own graphics over the top.

189

So the game engine only has to remember this

190

and the ai does that there's also another piece

191

if you ask for something the game rules don't cover a coding agent writes new rules into the engine

192

and updates the prompt the video model uses

193

so the world can change mid -game without the ai having to invent new physics on the fly down here

194

we can see three games a medieval castle shooter an ancient egypt shooter

195

and city racing these long play sessions can run up to about a minute

196

and a half without any sort of cuts now there are some catches each game has its own model

197

and each one is about five billion parameters each trained just for

198

that scene so it's not really a general make me any game model

199

and the output is 480 by 864

200

so it's not anywhere near hd now the project page does talk about running on a single rtx 5090

201

but the model card does say there's no guaranteed frame rate consumer gpu memory fit

202

or exact scene reconstruction so it may not totally fit into your gpu's memory

203

and fast camera moves and long sessions still can cause some drift

204

and the actual training code isn't included but you can get the weights

205

and the inference code under apache 2 .0 license

206

so you can use them commercially

207

now there is a hugging face space listed here for hugging apps matrix game 4 .0

208

so you should be able to play around with it here they do have some examples

209

that are pre -cached in

210

if you want to check out some of those examples

211

now personally i think this engine plus like ai paint job

212

or ai reskin is a pretty realistic path to ai games at least more

213

so than like asking a model to do everything turns out nvidia thinks

214

so too next up let's talk about something

215

that came out of nvidia called stream render h3

216

this is actually a pretty similar concept to what we just looked at it uses kind of

217

that same split code runs the world

218

and then a video model draws

219

it in real time the difference with this one is how

220

little the game actually hands the ai matrix game gives the model a blocky 3d render nvidia gives

221

it a color -coded map every pixel is a flat color

222

that means road or grass or barrier

223

or rival car there's no lighting no textures the video model actually draws everything you see

224

so here's the game engine's own view the color map that it actually sends

225

And then the final frame that, you know, the user sees now because the map carries no like looks at all,

226

you can actually restyle the whole game by swapping one reference image.

227

We can see here that this is all the exact same drive, but when we fast forward a little bit,

228

they can swap out the scene for anywhere they want and put the exact same drive in any different scene.

229

It's the same exact input.

230

What we see in the middle, just rendered out in different locations and different settings during the day

231

and things like that

232

so one thing to notice here this is kind of one of the hard parts for this model is

233

that the color map paints all of the rival cars the same exact color

234

so the model has to actually remember on its own

235

which car is which and make sure to keep the same paint job across the various cars

236

so a fun fact about this model is they actually handed

237

clod code the job of building a faster video decoder for this they gave it a speed target

238

and then they had a working version running in the pipeline in about three hours.

239

Now this does actually require some serious hardware.

240

Each rendering step takes about 218 milliseconds.

241

However, that's on eight GB 200 GPUs.

242

These are like the chips that they put into data centers from Nvidia.

243

So this isn't something you're going to be running on your gaming PC, at least not for a while.

244

Also, it's only been trained on racing

245

so far that's why all the demos are like race cars

246

so you can put this one kind of next to matrix game

247

which we just looked at

248

and you get a decent picture of the trade -off matrix

249

game aims for one consumer gpu at like lower resolution where nvidia is going for photo real

250

but on a rack of data setter hardware okay this one's a little weird

251

but i want to show it off because it's also kind of cool

252

so odyssey the world model company

253

that launched odyssey 3 not too long ago posted a clip saying

254

they put a fruit fly brain inside their world model

255

and now it flies around imagined worlds by itself

256

so this fly brain wasn't actually created by odyssey

257

it's a simulation built by a company called eon systems from

258

a complete map of a fruit flies brain about 140 ,000 neurons and 50 million connections.

259

Odyssey swapped out the virtual body that brain was hooked up to

260

and connected it to a camera moving through an Odyssey three world.

261

I don't know.

262

I thought this one was fun, but there's like no paper, no code, no explanation of how the two are connected.

263

And people in the comments did note that the fly never actually tries to land anywhere, which, you know, if it was a real fly,

264

it probably would, but it's brought up a lot of questions about you know simulation theory like does this fly believe it's real

265

and in a real world but it's actually the simulation and

266

if so what are the implications of round simulation theory for you know humans

267

and stuff we don't need to go down that rabbit hole though

268

while you can't actually play with this fly flying around inside of odyssey 3

269

you can actually mess around with odyssey 3 itself just head

270

over to experience .odyssey .systems give it a prompt of what you want it to create, and you can actually interact in this world.

271

I am steering this fox right now.

272

I'm telling it to jump.

273

I'm telling it which direction to go.

274

I'm changing camera directions.

275

All of this is me actually controlling this.

276

Pretty cool.

277

All right, moving into 3D.

278

Preferred Networks, which is based in Japan, just released Telescope, which turns a set of photos of a place into a detailed 3D mesh,

279

like from a single object all the way up to city blocks.

280

Now the usual way to get a high quality mesh from

281

photos is to spend a long time fitting a model to that one specific scene.

282

Telescope actually skips all that.

283

It's a generative model so it's learned what surfaces generally look like

284

and it fills in gaps where you didn't get many photos.

285

It also handles shiny and glass surfaces which normally wreck photo based 3D.

286

So the name is actually the method itself.

287

It builds the whole scene at low detail first, then zooms in on areas that need more detail and refines them like a telescope.

288

The colors here show which level each surface came from.

289

So if we look here, we have a Gaussian wrapping, which is sort of the standard method.

290

And if you look at this, it looks pretty good, but it's lacking a lot of detail.

291

Now, this is telescope and there's so much more detail here

292

so here's the original Gaussian wrapping

293

and here is telescope like it's a pretty big difference here's another example of Gaussian wrapping

294

and when I move this over you can see their version

295

and all of the details pop into focus

296

now there is a little bit of a catch

297

you do need a lot of photos to achieve

298

this you can see this one took 256 this one also took 256 right here

299

This one actually only took 56, which isn't too bad, and it can only rebuild what's inside the area your camera actually moved around.

300

This is one we don't have any sort of real access to.

301

It does say the code is coming soon, but it looks like a pretty promising leap from what we got out of Gaussian splatting.

302

Here's another pretty cool one called Gaussi Animate, and this one's actually used for animating 3D characters.

303

It's designed for rigging up characters that are soft on the outside

304

but rigid on the inside so people in loose clothes animals

305

or garments just on their own so a normal animation skeleton gives you control

306

but a loose shirt or like a dog's belly doesn't move like bones

307

so this adds on a whole bunch of other floating nodes

308

on top these additional floating nodes sort of capture all the jiggle

309

and flow and the additional physics that come along with the, you know, the cloth or the material on top of the bones.

310

So we can see here that it finds this sort of shape, but then once it finds the shape,

311

it actually puts a skeleton in the middle to figure out which parts need to actually be rigid.

312

Once it figures out the points that need to be rigid, then it figures out what points need to sort of flow outside of the bones.

313

So here's some other examples, like a clothed human, you have the skeleton, you've got the bones you've got the skinning

314

and then everything combined where you can see the the dress sort of flow

315

now you do need a captured motion sequence of the character first

316

it's not going to just work from like a single image

317

but the code for this one is available on github

318

and available for research use right now all right now this is the last one

319

and it's actually not ai but i just thought it was really cool

320

so i wanted to share it anyway this one's called world gasha i think I'm pronouncing that right.

321

And it's actually a free browser tool from a Japanese developer

322

that actually creates random fictional regions that has mountains and rivers

323

and coasts and towns and railways and race patties and almost anything you can imagine.

324

And it draws it as like this manga style line art.

325

Well, this one's actually not in that manga style line art.

326

This is a randomly generated city, but you can see if I add it with the lines, you kind of start to get that look.

327

You can also put it on lines and tone and you get like a black and white version.

328

You can put it on this like clay mode.

329

And this is again, a randomly generated seed.

330

I could reroll the dice and it's going to build me a whole new town or city or whatever.

331

So now I can zoom in on this city that it just created.

332

We can also change it to downtown.

333

It's going to build us a new city that looks like a downtown area or skyscrapers.

334

It'll build us a new city

335

and these are again all randomly generated every single time

336

and you could zoom into different views

337

so let's say we want like a top view

338

we can look straight down on this city like this maybe want an eye level view of central street

339

and it actually brings us into the city where we can look around

340

and wander around in this city now imagine using something like this

341

and then re -scanning it with like one of the ai models

342

we were looking at earlier you can build a pretty primitive looking city

343

and then use AI image or video models to reskin it into something really realistic

344

and maybe even build scenes for like films and animations and stuff.

345

Again, this is not actually an AI model.

346

There's no outside data.

347

It's just random numbers and math generating all of this.

348

And if you do like a specific scene, you can actually export it as a PNG or an SVG or even a 3D object file.

349

So you could theoretically create a world and then pull it into unreal engine

350

and build a game inside of it or something like

351

that i know i usually focus on ai stuff

352

but really you know the theme of my channel is just cool stuff

353

that i come across online that i want to share

354

and sometimes it's not always ai

355

but that's what i got for you today again

356

i make a news video every single friday where i break down all of the news

357

that came out in the ai world each week

358

but it's usually the biggest stories with the most impact on the most amount of people

359

i'm sort of trying to filter out the signal from the noise of everything that's going on

360

and help you understand what might actually change what you do on a day -to -day basis

361

and in doing that sometimes I miss some of these smaller research projects

362

that don't get a ton of love

363

and not a lot of people talk about

364

and I find all this stuff really cool and really fascinating and

365

if you go back like three years on my channel

366

this is the type of stuff

367

I used to talk about a lot more frequently until all of the big labs

368

and big companies started having new launches every single week

369

and I kind of shifted more focus to that

370

but I want to get back to my roots a little bit

371

and make these videos a little bit more often so

372

if you like videos like this where I sort of dig

373

and find some more underground things that are happening in the world of like AI

374

and just cool tech research let me know I'll make sure to make more videos like this

375

I'm planning on making at least one of these a month

376

because it's just really fun for me to dig in and find some of this stuff.

377

But again, that's what I got for you.

378

If you like videos like this and you also want to be super looped in

379

and informed on all the latest AI news, I'm going to keep up with all of that for you.

380

I put out a video every Friday with all of the news.

381

I also put out a newsletter twice a week.

382

You can subscribe for free over at futuretools .io.

383

But thanks so much for hanging out with me, nerding out with me today.

384

There's a lot to nerd out on i really really appreciate you

385

and hopefully i'll see you in the next one bye

App herunterladen

Alles, was du zum fließenden Sprechen brauchst

KI-AussprachebewertungJeder Satz bewertet

IPA-TrainingJeden Laut meistern

WortschatzBaue deinen Wortschatz auf

VokabelspielSpielend lernen

QR scannen

Download on theApp Store Get it onGoogle Play

FOR YOU

Empfohlenes Video

257 Sätze

How I use AI as a senior software engineer and 5 principles of AI explained in 10 minutes

Daily ConversationWorking Adults

102 Sätze

Run AI Models (LLMs) from USB Drive | No Filter, Fully Offline💀

Daily ConversationWorking Adults

455 Sätze

I Gaslit an AI Game Companion Until She Lost Her Mind

Daily ConversationWorking Adults

400 Sätze

I Made Every AI Play Detroit: Become Human

Daily ConversationWorking Adults

331 Sätze

Dear Higgsfield, I reverse engineered you in 12 hours

Daily ConversationUniversity StudentsWorking Adults

43 Sätze

Translating Claude’s thoughts into language

Daily ConversationUniversity StudentsWorking Adults

366 Sätze

ChatGPT + Podcasts = English Fluency? Try These 12 Prompts!

Daily ConversationUniversity StudentsWorking Adults

Wortschatz und Sprechhinweise zu dieser Lektion

Diese Sprechübung basiert auf dem Video „11 AI Breakthroughs You Didn’t Know Existed“. Dieses Video enthält 385 Sätze und 4734 Wörter zum Nachsprechen. Der gesprochene Teil dauert 25:16. Der Sprecher spricht schnell, etwa 187 Wörter pro Minute – rechne mit verbundenen und verkürzten Lauten. 84 % der Wörter gehören zu den 3.000 häufigsten im Englischen; den Rest solltest du dir vor dem Üben ansehen.

Wichtiger Wortschatz in diesem Video

15 lernenswerte Wörter aus dem Video, mit Aussprache und Bedeutung:

Wort	Aussprache	Bedeutung
generate Verb	/ˈd͡ʒɛn.ə.ɹeɪt/	generieren
odyssey Substantiv	/ˈɑ.dɪ.si/	Odyssee
audio Adjektiv	/ˈɔ.di.oʊ/	Audio-
telescope Substantiv	/ˈtɛl.əˌskoʊp/	Teleskop, Fernrohr
bone Substantiv	/boʊn/	Knochen, Gebeine
hug Verb	/hʌɡ/	sich umarmen
simulation Substantiv	/ˌsɪm.jəˈleɪ.ʃn̩/	Simulation
render Verb	/ˈɹɛn.dɚ/	zu etwas führen, machen
uncle Substantiv	/ˈʌŋ.kl̩/	Onkel
physics Substantiv	/ˈfɪz.ɪks/	Physik
hardware Substantiv	/ˈhɑɹdˌwɛɹ/	Eisenwaren
clip Substantiv	/klɪp/	Klips, Clips
wrap Verb	/ɹæp/	wickeln, einwickeln
demo Substantiv	/ˈdɛm.oʊ/	Demonstration
randomly Adverb	/ˈɹændm̩li/	zufällig

Phrasal Verbs, die du hören wirst

Wort	Aussprache	Bedeutung
figure out Verb		herausfinden
come across Verb		wirken, vorkommen
get back to Verb		sich melden
put out Verb		rausbringen, raussetzen
break down Verb		versagen, den Geist aufgeben
bring up Verb		erwähnen, mit etwas kommen / ankommen
check out Verb		überprüfen
come out Verb	/ˌkʌm ˈaʊt/	herauskommen

Sätze zum Wiederholen

Kurze, vollständige Sätze aus dem Video, die du im Alltag wiederverwenden kannst:

Then why is your coat still wet?
Nephew, why you put ketchup in egg fried rice?
Did you catch that?
We're getting closer and closer.
What do you think?

Grammatik in diesem Video

Die Strukturen, die der Sprecher am häufigsten verwendet, mit den genauen Worten aus dem Video:

Struktur	Im Video
Present Perfect Continuous have/has been + -ing – eine Handlung, die früher begann und noch andauert	i've been using
Present Perfect have/has + Partizip Perfekt – eine vergangene Handlung, die jetzt noch wichtig ist	I've come · you've ever watched · have reported
Passiv be + Partizip Perfekt – wichtig ist, was geschieht, nicht wer es tut	be desired · being fooled · was filmed
Relativsätze who / which + Satz – eine Zusatzinformation über eine Person oder Sache	page, which is · model which is · people who recorded

Aussprache, auf die du achten solltest

Der Sprecher verwendet 69 Kurzformen und abgeschwächte Formen, zum Beispiel I'm, don't, you're. Sprich sie in der kurzen Form, so wie du sie hörst.

Die „th“-Laute: mathematics /mæθ(.ə)ˈmæt.ɪks/, cloth /klɒθ/
Die Laute „sh“ und „zh“: simulation /ˌsɪm.jəˈleɪ.ʃn̩/, capture /ˈkæp.(t)ʃɚ/, shooter /ˈʃuːtəː/, reconstruction /ˌɹiːkənˈstɹʌkʃən/, mesh /mɛʃ/
Lange Wörter – auf die Betonung achten: reconstruction /ˌɹiːkənˈstɹʌkʃən/, explanation /ˌɛkspləˈneɪʃən/, ordinary /ˈɔɹdɪnɛɹi/, hilarious /hɪˈlɛə.ɹi.əs/, artificial /ˌɑː.tɪˈfɪʃ.əl/

Laute, die Deutschsprachigen schwerfallen:

/w/ — mit runden Lippen, nicht wie das deutsche „w“ (/v/): hardware /ˈhɑɹdˌwɛɹ/, sequence /ˈsiː.kwəns/, quietly /ˈkwaɪ.ɪt.li/, swap /ˈswɑp/
Stimmhafter Auslaut — /b/, /d/, /g/, /z/, /v/ am Wortende nicht verhärten: lab /læːb/, hug /hʌɡ/, rigid /ˈɹɪd͡ʒ.ɪd/, cloud /ˈklaʊ̯d/, footage /ˈfʊtɪd͡ʒ/
/æ/ — offener als „ä“: lab /læːb/, wrap /ɹæp/, randomly /ˈɹændm̩li/, capture /ˈkæp.(t)ʃɚ/, angle /ˈæŋ.ɡəl/

So übst du mit diesem Video

Höre dir das ganze Video einmal an, ohne zu sprechen, und notiere die Wörter, die du nicht kennst.
Beginne mit 0,75-facher Geschwindigkeit, sprich Satz für Satz nach und wechsle zur normalen Geschwindigkeit, sobald es leichtfällt.
Nimm dich auf und vergleiche mit dem Original; achte dabei auf Wörter wie generate, odyssey, audio.

Was ist die Shadowing-Technik?

Shadowing ist eine wissenschaftlich fundierte Sprachlerntechnik, die ursprünglich für die professionelle Dolmetscherausbildung entwickelt und durch den Polyglotten Dr. Alexander Arguelles populär gemacht wurde. Die Methode ist einfach aber wirkungsvoll: Du hörst englisches Audio von Muttersprachlern und wiederholst es sofort laut — wie ein Schatten, der dem Sprecher mit nur 1–2 Sekunden Verzögerung folgt. Anders als passives Hören oder Grammatikübungen zwingt Shadowing dein Gehirn und deine Mundmuskulatur, gleichzeitig echte Sprachmuster zu verarbeiten und zu reproduzieren. Studien zeigen, dass es Aussprachegenauigkeit, Intonation, Rhythmus, verbundene Sprache, Hörverständnis und Sprechflüssigkeit signifikant verbessert — was es zu einer der effektivsten Methoden für die IELTS Speaking-Vorbereitung und reale englische Kommunikation macht.

Shadowing-Technik: die vollständige Schritt-für-Schritt-Anleitung lesen →

Shadowing-Übung: 11 AI Breakthroughs You Didn’t Know Existed - Englisch Sprechen Lernen mit Video

Übungsmodus wählen

App herunterladen

Is AI About to Change Movies Forever?

8 THINGS EVERY LANGUAGE LEARNER SHOULD DO DAILY

Gemma 4 Just Released - Whats In? In 2026!

Jack Dorsey’s New Dessa AI Voice Sounds Scary Human

9 Insane Claude Mods I Already Can't live Without

How I use AI as a senior software engineer and 5 principles of AI explained in 10 minutes

Run AI Models (LLMs) from USB Drive | No Filter, Fully Offline💀

I Gaslit an AI Game Companion Until She Lost Her Mind

I Made Every AI Play Detroit: Become Human

Dear Higgsfield, I reverse engineered you in 12 hours

Translating Claude’s thoughts into language

ChatGPT + Podcasts = English Fluency? Try These 12 Prompts!

Wortschatz und Sprechhinweise zu dieser Lektion

Wichtiger Wortschatz in diesem Video

Phrasal Verbs, die du hören wirst

Sätze zum Wiederholen

Grammatik in diesem Video

Aussprache, auf die du achten solltest

So übst du mit diesem Video

Was ist die Shadowing-Technik?