Mindbending Text to Video AI is Getting Better Every Day!

[@Canal10000]

Pika is definitely the best I have seen so far. Cant wait until next year when we start to get some pretty good short videos!

[@legacylee]

Man, the gens I've been getting has been pretty good. Funny those drone shots were actually my generations, and it was supposed to be drone shot of the Thundercats standing in front of their Lair, but you I got lil drone like thingy's hahah. My Lion-0 generation though, that was pretty top notch. I see a few of my generation in here, I'll have to give thanks to AI Flux for dropping them in this video. I feel special now haha

[@Pawnsappsee]

Matt always bringing the best AI stuff, I already follow 3 Matts on youtube and each one of them never let me down.

[@martiddy]

3 Matts? I only know this channel and Matt Wolfe

[@theyoutuber1200]

dangmattsmith perhaps? idk

[@techiechar]

Wait, we’re missing a Matt?

[@80sGameGuy]

I follow Mattpat from Game Theory and Matt Tye aka laowhy86

[@weluvtech]

Hi Matt (YT deleted my comment 5 hours ago from my mobile, so I'm on my PC now), thanks for showcasing my FullJourney videos. 😊 I did all those Beatle clips, and the wild African animals walking around the streets of Sydney (I'm actually in Adelaide) lol. FJ can be either very coherent, or very incoherent. Either with very short, or very long descriptive prompts. So hence why I repeated many of my prompts several times, to get some good clips to use on my own channel. I've also signed up for Pika, and waiting for approval. Did you get in already? Thanks.

[@aresaurelian]

Once voice instruction to video reach 3 minute scenes, anyone in the world can create full feature films.

[@hombacom]

How can random creepy scenes become a movie that make sense

[@BastianAI]

coo, that's my prompt at 1:47 . i was just thinking about the time i found something really nasty in my flour which made me sick lol

[@racksityentertainment]

AI really doing better than my brain creating videos!!! My dreams are lower quality than this 😮 … imagine in 10years . AI producing full HD consistent videos !

[@XsiadzBiskup]

Stable Diffusion is a lot better than the first model, if you use the tokyojab method.

[@nelsonporto]

Loved to see my surfing chickens getting famous! 😃😃😃

[@mohammedrashid1910]

cant wait for this to hook up with ControlNet and start off from a single image. would definitely be better

[@MeNoOther]

If people made storyboards, which can be a few seconds to minutes, than edit them together. That's a movie! There should be an ai generated movie script, then take the scenes and make the shots.

[@ChessMasterNate]

What should be interesting is the dynamic between ordinary moviemakers and generative moviemakers in the future. Will there be conflict, or will they just merge. Though, going by the science fiction shots I see on Dust, the visuals will outclass the plots. Unless, AI helps them out there as well. I did lightly test plot stuff. Seemed very cliché. That is okay. I am good at plots. The only plots that seem plentiful today are conspiracy theories. That is sad. I don't think anything is out of reach for AI. So, I still hold out hope for great plot generation in the future.

[@creatorsmafia]

The future of text-to-video AI is here, and it is getting better every day! Pika Labs and ZeroScope are raising the bar with their impressive capabilities.

The Alien DJ reminded me about Imphenzia when he made music, I think he still use an Alien as logo.

[@arielcurra7647]

The aliens were really good

[@samuelhau666]

Thanks for the info

[@TheRedisReal]

I think I'm first? but either way I'm very exited for the future of AI

[@agnesslovehealz]

Like the music 🎶 touch

[@agnesslovehealz]

Zeroscope 3.0 looks promising

[@misteraiguy]

Good Score I like them both!

[@frost_spirit3314]

This feels like the early dalle models

[@misteraiguy]

Was just looking for some other options to use rather than GEN2 which is great but I want Variety

[@MeNoOther]

You broke the discord server! LoL

[@Interdimensionaltravelagent]

Been working with Gen2 for a while. Still nothing quite as good if you use it right, but getting there.

[@BOBLAF88]

AI is a fun time! We are truly at the prelude of some mind altering 💩🤔

[@BTMYYY]

I'm excited that's for sure 😀

[@BTMYYY]

i wish i had access to it locally

[@T.E.S.Channel]

Best text to video = Kaiber. By far.

[@dalecorne3869]

The link for Zeroscope says the invite may have expired or you might not have permission to join.

[@arielcurra7647]

This is great!

[@legacylee]

[@dreamphoenix]

Thank you.

[@sinayagubi8805]

Bro, you didn't give the opensource one enough credit. it had not a single nightmare fuel unless where it was asked for it. so it's actually ground breaking. the first one was meh.

[@Xiiboca]

Ashuashua pika

[@WilliamAlanHarris]

Link is expired for zeroscope v3

[@Metasandwich2]

The shit Discord AI free server doesnt work, ones have to wait like forever for one single 10 sec video to be generated

[@mmcool]

Why not AIs training AI? I mean you can let few of them talking to each other very fast an you build in short time a very mighty Modell

[@hombacom]

So how they learn something new and confirm it’s great for humans

[@zenmasterjay1]

Barely a couple years and u will be able to live up to your name😂

[@legacylee]

I got invited to FJ too hah

[@КошНаран]

😮😮😮

[@Yozo_official]

8h 637 like 202k subs

[@WelshDog]

👏👏🔥👌👊

[@danielmartinmonge4054]

"video generation has been too slow so far" What?! Bro! This wasn't even a thing one year ago! How fast do you want It to be??!! Hahaha

[@KuldipTheBlueMoon]

Please help me for Sign-up or give me a user name and password to log in. I am fed up.

[@pretornad729]

runwayml is such a joke LOL they, this is 1000x better

[@igorbessa3563]

Video generation is light years behind image generation. I think it will take at least 10 years for it to catch up

[@bambinoesu]

i don't know if its just me, but gen-2 text 2 video is poor. it was pretty much just generating still images with very minimal movement

[@SingularityReacts505]

Hey Matt Vid Pro I want to extend my cinematography experience to your videos i can guide you on how to light really easy and ill do it for free let me know brother.

[@MattVidPro]

I know how to light my room is just horrible for it. I have my desk where I want it and frankly that is more important than lighting for the moment. I am going to buy tint for my window so that it exposes properly on camera

[@SingularityReacts505]

@@MattVidPro of course I hope you have a wonderful day and thanks for all the ai news it means a lot to the community, everyday I look forward to watching one of your videos .

[@supfiger]

Hi. Change the place of shooting. Bye.

[@HeedReactionzz]

Firsr

[@krispeekornflex]

you need sleep.

[@ytubeanon]

3:13 looks like something I might see in a dream where my brain is trying to form a narrative out of random stuff... most of it all looks like Nightmare Fuel AI

[@Ben_D.]

One year from now, this is going to be much better. I remember playing with the first Dall*e model and thinking it was both impressive and utter crap at the same time. This is kind of that. One year. It will be fine. 😏

[@Pawnsappsee]

How long will it take to generate an entire anime opening with a prompt ?😅😂❤

[@gongontonato]

​@@Pawnsappsee year and a half maybe?

[@hombacom]

It’s still one-hit wonder scenes that doesn’t fit together just like ai images don’t

[@gongontonato]

@@hombacom incorrect. you can now create different images with the same character in different postures accurately now. Only thing left is to maintain an consistent background scene and a bit of polishing and BOOM.

[@hombacom]

@@gongontonato Yeah you just press a button and BOOM. Different postures don't mean it look good in different scenes. AI images have a lot of details that don't make sense. But lazy people can still dream.

[@reezlaw]

State-of-the-art text to video is going to be the real revolution in entertainment

[@awesomebearaudiobooks]

I think you haven't included the FullJourney discord link into your video description.

[@moustachecat1658]

:)

[@virtual_intel]

Not free it’s a no go for me long term

[@vvolfflovv]

Something tells me 95% of us will be spending the majority of our time in the NSFW tab :)

[@sophiacristina]

As a brazilian, i'm not sure if i want be caught handling Pika... :s

[@metakron]

2022: text to image development 2023: text to video development

[@VaibhavShewale]

this open source one is cool, also audio in video damn!

[@geralpink3014]

Greetings from Puerto Rico 🇵🇷

[@BhernardoMocambique]

I love that in my language (brazilian portuguese) pika is a slang way to say penis

[@ai_vids]

You are the best AI divulgator on YT ❤

[@hipjoeroflmto4764]

Dada

[@womenwelove]

honey, you need to sleep more...seriously...take a break...

[@womenwelove]

@@SefSufi you are weird...

[@PerfectArmonic]

See them after one year…

[@ReluctantStallion]

Looks awful.

[@sinayagubi8805]

I would rather pay for Zero Scope than Pico Labs

[@parsley8188]

This is really cool!!

[@CharlesMcCarthyUFC]

Hey thanks so much for the coverage!

[@1EAS1World]

gosh the servers so nsfw lmao

[@FrotLopOfficial]

This + text to voice. Oh man political hit pieces are gonna get nuts

[@The_Questionaut]

Misinformation is going to be crazy

[@FrotLopOfficial]

@@The_Questionaut Yea man the 2024 elections are gonna be world changing. I cant wait lol

[@FusionDeveloper]

cool

[@Nutronic]

That music over the generations is annoying.

[@jeimycalvo6337]

Matt lo amo a usted usted