#flow-general・🌊
1 messages · Page 15 of 1
It just seems like it's almost pointed that I will just not be able to use the services that I paid for. What's even more frustrating if the fact that there's absolutely no way to access any kind of customer service after you drop two or three hundred bucks
They just make you post on a forum and hope somebody else is probably in the same situation and replies I guess?
Personally I'm finding almost no reason to use Veo2 or Veo3 anymore when Veo3 fast is so good. In fact the last head to head tests I've run of Fast to Veo3 quality, Fast has actually been superior.
This is how simple I made it and it still couldn't work
😆 this is why I don't like Gemini
On flow I only see an option for veo3 which is a beta
Cant seem to get it to make the hummingbirds wings flap like it should.
AI video still struggles a lot with high-speed articulation so like things like hummingbird wings that are moving at an accelerated rate of speed are very difficult to render even without AI
Any tips on how I can get it done?
did anyones audio stop working?
Yes I tried to create the same 5 five times in Veo 3 and audio failed every time
This is shown when ur limit is reached
Which is weird because I still have over 10,000 credits left so my limit isn't reached and it started showing it yesterday the day that I bought the ultra plan
No no in gemini
You mind sharing the prompt?
In gemini u can create without credits
I think this intro part is what kind of encouraged the wing movement:
Prompt: "Hyper-realistic slow-motion video of hummingbirds with ultra-fast, blurred wings (50-80 flaps per second)
Thanks I'll try to incorporate that!
Anytime
Yeah it'll do one a day but I think there is a limit to the depth of your prompt
Guys how can we use flow on mobile
I have not used Gemini at all today to create anything and it still won't let me do anything and I think I was able to create one video on Gemini yesterday
Web browser
It resets according to the time when was ur limit reached
Oh trust me I tried that. I have a surface tablet on my counter and I opened Chrome because in brave browser the video button doesn't even show up on gemini. Then I use it on my custom desktop and same thing no audio nothing
Audio comes eventually
Okay now can I ask what’s a custom desktop 😭😭
I built custom gaming computers. So when I say custom that's because I built it it's not something I bought off the shelf
What prompt are you doing for the accent
But after 1 hour I checked it got audio automatically
What's the point of eventually with a daily limit?
I've tried prompting british and australian.. i get nothin
I ma talking about flow.
I actually have no idea. The first video had no accent in the second video did 😆
yep
I can look through my props and see what it exactly was but as far as I know I've never asked for an accent it just kind of stuck it in there
just the random nature of it
I'd love to have my canadian sasquatch sound quebecois and the mexican one sound mexican
Looking at videos when this started happening late last night they still don't have any audio and videos I created this morning about 12 hours ago also don't have audio so I'm not sure what you're getting but I wish I could at least get that
I've sort of noticed that peak hours it gets pretty back
Blanco the yeti or something like that there's one on tick tock and sounds Spanish and speak Spanish and it's hilarious
bad
This guy
Oh damn!! Understood
Maybe this is off the Mark today but I think the prompt was speaks anti-ice 😆 sorry I couldn't resist
I'ma try speaks like a cholo though
it's more of a persona so it might work
Scene description: TV interview at a busy open-air market in Manaus during a power outage. The camera wobbles slightly as the crew navigates through dense crowds. People shout in the background, trying to keep their stalls lit with flashlights and candles. Smoke rises from a nearby food stall. In the middle of the confusion, a tall, shimmering crystal person stands perfectly still, refracting the candlelight in prismatic patterns. Occasional chimes ring softly as the wind brushes across their crystalline surface.
Interviewer asks:
"Como esse apagão afetou sua rotina aqui no mercado?"
Interviewee answers:
"Sem luz elétrica, meu corpo começou a absorver segredos esquecidos nas sombras antigas... tlim."
All that fit in 8 seconds? that's a lot of words
even if you put do not use captions, do not add music and do not add subtitles.. they still randomly appear
drive me mad
I've noticed on a few of my riot jokes, the ai adds breaking glass at the punchline instead of a rimshot
i swear 90% of my credits go to just trying to get the right timing and intonation on my jokes
i wish next big update veo will fix the issue voice glitch, hand glitch, letter glitch,
veo 3 generate more cartoon video than veo 2. kinda annoying..
Does anyone else have the experience of only the first generation being really good, everything else you prompt is not so good
thats me, quality getting worse and worse after first generation.
same here
it fells so rough around the edges
we need more options for control, putting the same prompt in twice still gives vastly different results
QUESTION FOR FILMMAKERS: I am looking for a way to have a consistent 4K input and output across platforms and was wondering if anyone had any knowledge of how to do this. Here's my problem: I am making at Generative AI film that needs to be 4K ready for festivals. I am making the still images in MidJourney and choosing to "Upscale Subtle" for the final image. They look great! Then, I bring them into Photoshop to upscale them using the Resample: Preserve Details 2.0, which I find better than Gigapixel. There, I upscale them again to 3840x2160 for 16:9 standard 4K resolution. I bring this into Flow for animation and it outputs to 720dpi. I could upscale internally to 1080 but I find it really loses a lot and creates unnecessary contrast etc. So, I bring them in Topaz and upscale with NYX, which is best for my types of images and output them to 4K. So, the problem is that I go from Photoshop outputting at 4K, then back down to 720 output in Flow, then back up to 4K output from Topaz. I lose alot in the "Flow" stage. Does someone know a work around for this? FLOW CREATORS: is there a way to get 4K output by working with professional filmmakers? Thank you!
youre just gonna have to wait until theres an AI model that outputs at 4k resolution
The results so far aren't terrible, and could probably pass, but waiting for the big breakthrough sucks.
If you really are using it professionally I would suggest upscaling the images via Magnific AI with the creative upscaling, as it brings a LOT more detail than just the subtle upscale from MJ and photoshop upscale
#intros・👋 Hi everyone I'm Heather, new to this group. Not exactly sure what I'm supposed to do though. ?? Anyone have some advice?
Hi Heather. Just got here myself. Looking for 4K answers...
so skip the upscale in Midjourney and take the lower res image right to Magnific AI?
How can I remove captions for my Veo 3 video generations? Ive tried adding "Do not add captions" to my prompt but that does not work.
Quick question. Can you change the aspect ratio in flow?
i heard that using quotations does that
try just to use semi colons or brackets
i also think you gotta ask it to generate in 9:16
ive tried asking for 9:16 for vertical flow output many times. it rarely works. Or, it will present as 16:9, then put a vertical image in the middle of it...
FLOW needs two things desperately: 4K output. 9x16 Vertical option.
Yes
It looks so much better
But it is an expensive process
how to make more realistic video .sometime video generate cartoon video ...
Id be happy with consistent audio and continuity.
@jolly portal Hi, nice to meet you. Hope you find your answers. Seems like your grasping it better than me so far. 💯
thank you
Hey, what are you asking for?
this here has flow and can do it in 9:16, so ive been told: https://www.aicut.pro/
https://www.aicut.pro/ in here you can use flow in 9:16 https://www.aicut.pro/
4K input (MJ or Topaz) → Flow → Topaz upscale. It won’t truly give you native 4K, but it will deliver the cleanest possible result. Wait for industry advances for true 4K.
You could also just use Luma Modify, it is pretty good to modify the aspect ratio of videos
really? interesting. i'll check it out, thanks!
Hi all, I've a Google AI Subscription but Flow doesn'twork (it says I haven't). Any clue?
PRO TIP, FLOW TV has all the prompts for really good ideas that can be easily swapped out with your own ideas using ai!!!
Hello, I'm Frank, Co-Founder of CreatorX, a Web3 platform for creator asset issuance and trading. We currently have 10,000 seed users and are looking to integrate AI tools to empower creators. Are there any collaboration opportunities?
Hey! I have a problem with Flow — I already subscribed, but it still says "You must be a Google AI subscriber to generate."
When I try to subscribe again, it says "You’re already subscribed."
Any idea how to fix this?
this has been a problem since release
I don't know why this is not fixed yet
this is pure tax, 16 credits per video and 30 credits costs 18$
So any workaround? How other people do?
img to video on whisk. add a sound would be cool
yo i have subscribed to gemini pro but flow still asks me to upgrade and after clicking on upgrade
they are from eligible countries
I'm starting a project called singing history. Working with Suno, Gemini, and Flow. My plan is just to feed historical documents in and work to turn them into songs.
man this cheaper veo3 is a cheat code. now all we need is 9:16 ratio
then Veo 3 is the final boss
how do I make the videos longer then 8 seconds
cant
Oh so I gotta make multiple videos into one long vid
yessir!
Okay thank you so much
just subscribed to google ai to use flow, but it says I'm not subscribed? Please help
is flow ai credits are not refilling ?
why are some of my prompts not adding audio ?
Hi - new here, I am curious I signed up to veo3.ai thinking that would be the best place to create this videos and it seems i was mistaken that getting a google ultimate AI account via flow because it uses veo3 is way better. Is this correct? I have been having huge issues with veo3 and getting it to output what i am putting in the prompts even though the prompts i am using are generated from openai gpt for veo3. I dont want to get the upgrade on flow if i will have the same issues ... any insight would be appreciated
flow doesn't really work with veo 3
it works great with veo 2 but the frames to video feature on veo3 is very poor
What can we use then
idk trying a mix of veo-2 and veo-3 atm
100 credits for 1 veo 3 quality is jsut insane tbh
agreed
got it - so would you say that veo3 was a waste of $$ the yearly was like 500 and it just doesnt work on the prompts
well u can just try it out theres a free trial
not sure yet veo-2 + veo-3 does a decent job atm so idk about waste of money, tbd on that
@rancid bison
just subscribed to google ai to use flow, but it says I'm not subscribed? Please help
Hi all, I have the free trial version, but the app itself is pretty costly and I'm worried that it'll auto-renew. Looking at my billing page, Flow is not shown here is it hidden somewhere?
If you google Veo2 or 3, you will find a number of other sites that use it, I wonder if they are better and cheaper than google/flow. and would be nice if less restriction too.
anyone?
wow
:/
do you have like in your dialogues that kinda mature/sexy in nature... i feel like the AI is kinda tricking us sometimes to allow the generation, to waste your credit, but mute the sound if the dialogue is kinda mature or sexy. that happened to me a few times.
it depends on which veo version you use - not all of them have audio i think its veo2 fast that has the audio on flow
on an iphone you can just download to your photos and edit/crop. easy peasy. Takes 2 seconds.
yeah i do it in capcut but still cuts part of it off, not a good as if it was generated in 9:16
seems kinda pricey for the outcome when the results kinda suck and dont even always have audio. out of the last hour i had 3 with audio and the rest wouldnt generate any wasting all my credits for my trial.
this was the problem, lol i wasted like 300 credits
GG's
Who knows VEO 3 prompting rlly good so you get the same output every time? I need some help please
that would be sweet fr
but quality veo 2 is also 100 credits. also veo 3 on flow works good for me
there is a youtube video online. search veo 3 monkey/jeti prompts
Honestly there is this Toli the Trencher videos a guy makes which are amazing
I'm trying to learn how he makes them, what kind of prompting he uses
السلام عليكم
the credits restart every month?
yeah I use veo 2 fast which is 10 only
yeah, if u want wo sound its good. but if you want more quality, veo 3 still works better for me. luckily veo 3 wo sound is only 20 credits
the credits restart every month?
I think so yes
keep getting bug that it says i need a subcription whilst i am already subbed
just subscribed to google ai to use flow, but it says I'm not subscribed? Please help
same issue here
its a general issue but not fixed apparently
damn we need a vertical format asap
hey guys, is there a possibility to host in our server the veo 3 fast type API please as Flow propose ? @hybrid helm @rancid bison I am trying to build an app, and that would be really helpful to get this API source directly from you!
"jump to" scene builder not working for veo3? it's forcing me on Veo 2?
veo3 is good for realistic outputs. prompts should be more specific. when you use chat gpt add words like make it as much detailed as possible
you can kinda do some things with ai studio.
think it's still veo2 based though
thanks for the answer! like ai studio is providing api access?
oh lovely okay, I am interesting for veo3 tho since this is the main interest atm lol
even flow downgades most of what you would do with it in an api. I think....since only text to video on the first 8 seconds is veo3 ... well veo3 with just sfx after in flow
(first frame)
alright, thanks then, so for you veo2 can be as powerful as veo3?
not really. I shifted my plans to do audio myself.
just using it for video at this point.
doubt I keep ultra past this month.
will certainly keep pro
oh interesting, you using eleven labs ?
you think you getting better results ?
good ideas
It's more the limitations. Actually, I'm thinking of just using dubbing ai voices as long as I avoid people voices or those that may be otherwised copyrighted I have use rights...so just change my voice and do my own prompts.
Hello, i have a little problem where i paid for my subscription but i still cant access Flow AI, any help ?
did you find a solution maybe or just canceled your sub ?
An elderly Moroccan woman wearing modern and stylish clothes (like jeans and a fashionable jacket), with dyed hair and sunglasses, standing in a Moroccan street, smiling playfully and saying: "Aji Ḥamza, rah bghak ljwajel" (Come, Hamza, someone wants to marry you). Her facial expression is humorous and warm.
Did you guys figure out the audio issue?
@blazing mortar
sorry bro i didnt figure it out since i still cant access the flow in a first place xD
Can't access flow yet I'm a pro 2 terrabyte user
Same 🤷🏻♂️
This is how gemini is responding when I try to make a video right now.. I cannot create videos, but I can generate images and text. My previous response was a detailed description of the scene you requested, designed to evoke a cinematic feel.
в этой нейронке бесплатно можно генерировать видео?
When will it be available in EU?
why the extensions dont have audio? trash
Gotcha; thanks 🙏
So are any moderators going to help <@&1173646484311384114> with the no audio, a bunch of use have lost credits and our videos don’t work for a service we are paying for.
Hi , how to share a video from Flow website ?
like this " https://g.co/gemini/share/xxxxxxxxxxxx "
The moderators are not here for product support. You can check out #start-here・📍 for a guide to getting assistance. 😺
I'm making a video with a caveman speaking broken English. Every single video is generated with a British accent. How can I get it to generate an American accent?
try changing your prompt. but really you can't control the voice much at all
if you want a particular voice you'll need to use another program and dub it post
try telling it the american caveman, says
i doubt that will work though
I've tried all sorts of things
I'm just happy when words come out of the right holes at the right times 😛
because veo 3 is new.. it works fine for veo 2.
when you do it you do see the big words BETA next to veo 3 when you select? LOL
specifically says for audio...
probably mean audio is dodgy at best...
chkon li ando chi hel ndiro les compte bzff dyal usa
Quick question, I have google AI Pro but on Flow it says that i have to be a subscriber. If i press on that message, it brings me to google one and says "You are already a Subscriber". how can i fix this problem?
How are people making consistent scenes? Because for me the extend scene feature glitches and doesn't include audio
Whats the prompt?
Hey, I am using flow. how do I get my videos to display in portrait orientation? The default is landscape. Please let me know
Is there any test server for flow?
I do it with consistent and very descriptive prompts. Usually written by chatgpt
even then it's usually a miss, but 1 in 5 is generally pretty close
Best app to stitch multi videos together ?
i've been using capcut
Prompt is: "A cartoonish caveman with a wild beard sits in a dim stone cave, looking frustrated at a glowing stone tablet showing a volatile interest rate graph. The walls are covered with crude drawings of dollar signs and arrows. His stone wallet rattles. He grunts and says, “Why borrow with floaty rate? It go up, me cry.” The environment should look primitively financial — think Flintstones meets DeFi. Emphasize humor and confusion."
Need something mobile friendly too
Okay the app itself got some good tools? I just got everything set up so overwhelmed with all the tech lol I’ll check out flow a bit more
you're gonna go blind working on that small of a screen..
yeah I’m just on a flight I wanted to get busy with it now lol
Has google published any kind of project plan or update plan for upcoming updates? Or do they do that all silently
That's what I figured
I pay $125/mo.. they should treat me like a gamer and tease me with whats coming
has anyone had any good results with geolocation? I did a clip with a local pizza shop in it. It wasn't quite right, but it knew enough about the location to make it pretty cool.
I wonder if it changes the details just a little bit on purpose
mutliple clip attempts.. same prompt.. consistent on the characters. still hit or miss on the voice but getting there. Dialog still drives me crazy.
I added plain to the body armor description. it likes to add words
I wish the flow site would let me store character prompts so i could quickly cut and paste them or assign them to a function key..
I'm getting there.. Just need the dialog to land right
or should i say, "another dollar bites the dust!"
I guess it's teaser time.
What page is that?
hello I started using flow veo 3 today I made some videos they all had audio but for some reason it stopped working when I insert some talking text in prompt audio is silent there is no sound anyone knows how to fix this ? Wasted a lot of credits 😦
where do i prompt guys? on discord or on website?
can someone help me with veo3
Yes, but only if you're more specific with your questions. Because "Can someone help me about this thing regarding everything about it" is pretty low effort 😛
My prompts keep failing me. I try with Gemini and ChatGPT but can’t get it right. Always details wrong
@charred sail would love your help if you can assist me. I see you have been successful
There's always going to be a random element
This is hilarious... but it's not immediately clear what side you're on
I'm on the side of comedy
Oh you're one of THOSE 🙄
I'm a standup comedian.. so yeah 😛
brb checking Sid Jenkins
Call me crazy but I think you're actually a pro comic. I do not believe you're a fictional character
LOL nope i'm real
Label one of your skits, I'll guess who you are 😄
How do i get AI credits
I stand on the corner in my dirtiest clothing with a sign.. "need cash for AI"
Brilliant. But this skit is not yet published! That's what I said
sasquatch can do magic apparently
bruh ;c
I mean if any side can it's the educated side
can someone help me acces it i sadly cant because im in germany
I write them all kinda dumb cuz they act that way
BOOM! I found the side! Agreed
sounds like you need to rent a hosted server in the US
I don't have a side
Nobody i have ever voted for has been elected.. ever
i also need an us gmail or they dont let me use it
You voted for Yang last huh
my voice doesn't matter in any way except to get laughs apparently.. so that's what i do
I have 1 rule. Don't be mean.
can anyone provide us gmail ?
Bro these bots
not a bot btw
Explain
i need an us gmail because i cant acces it from germany
which part? magic sasquatch?
reis iyi aksamlar did not expect a turk with his real name in this chat
LOL that's a weird thing to care about
iyi aksamlar arkadas
naber amerika'da hava nasil
Hi guys, how do I get consistency of the style of the video?
hahahaha
well....
yeah, that's what I'm doing but that 720/1080 dip is killing me. I wish it was 4K PSD to 4K flow to 4K Topaz if needed.
you can try.. it's not easy and veo 3 loves to add random stuff. the extend and frame capture functions on Flow don't really work all that well right now so... you can try
Does Topaz do video now? Is there no way to upscale 1080?
1080 does not qualify for festival runs outside the AI animation community (like festivals for the Oscar runs etc). Also, please read my post, I already mentioned that about Topaz.
So at the moment I just get what I been giving ?
If anyone nails down multi character dialog.. let me know please.. this is killing me
I just try to be as detailed and subject oriented as possible.
I get pretty consistent results on some settings.. still trying to get a comedy club built right
is there quality difference between veo 3 fast and veo3 qualty?
Oof... Give it a couple years
Yes. Fast clearly has more quality than quality
still getting multiple errors on flow e.g. "You have to upgrade your subsciption plan to work with Flow" and when clicking on upgrade it says "You're already subscribed. You can access your membership benefits anytime." anyone else got this problem?
eventually i get one that's perfect. I just keep using the same prompt..
Are you clicking that little Reuse button or manually pasting it?
good as it gets i guess
I do both. depends
I suspect the Reuse button is not a blank slate
wait, why is the squatch wearing tactical gear.. /sigh
just random stuff like that
got credits for 3 more tries LOL
sasquatch scene generate with veo 3 fast?
This is how i set it up: A realistic, cinematic 8-second shot, without captions, without subtitles:
Sam is a Yeti from Minnesota. He has thick, shaggy white fur, with a slightly darker grayish-blue face, giving him a rugged yet noble appearance. His deep-set, intelligent eyes glow faintly with an icy blue hue, and his brow is heavy but expressive, hinting at both wisdom and strength.
Sam wears full police riot gear—sleek plain black body armor with reinforced padding over his broad chest and limbs, a tactical utility belt, and riot-resistant gloves. His gear is slightly scuffed from battle, showing signs of use.
Juan is a Sasquatch from Jalisco. He has thick, shaggy brown fur with a lighter tan-colored face, expressive amber eyes, and a wise, noble demeanor. His towering frame is massive yet dignified. He wears a flowing Mexican flag as a cape, draped proudly across his shoulders, contrasting vibrantly with the natural earth tones of his fur. His fur is detailed and textured, with a rugged, weather-worn look.
I was hoping by putting in cities they might speak like they'e from there, but nope
2 minutes or so.. not long
sorry for make you confuse, is sasquach scene generate with which option usually generate with. veo3 fast option or veo3 quality option?
what program is better to remove veo watermark. ultra membership. is too expensive..
is ur audio working?
for the most part.. sometimes it doesn't
are you using veo 3 fast?
no i use quality
I have no idea why he runs in like that.. in the prompt he's sitting next to him
I think this is the best its gonna get
I’ve just started a new channel focused on Bigfoot content. Hoping to enjoy the process and grow in this niche! I can’t decide — should I combine short clips into Vlog-style videos, or just keep making 8-second Shorts? I'm still working on character and environment consistency. That's why I don't want to waste my credits. Voice consistency was the easiest part. I really hope the Veo3 developers consider important issues like character consistency.
i have the announcment "Flow is not available in your country yet." im living in austria why i have that?
hey fam wen the credits reach 0, how much i need to wait for it to recharge?
change upgrade plan, invite friend to different account, use student plan, wait another next month
wach kayen chi mghribi hna nhadro
It's monthly.
I use this prompt, korean guy in los angeles korea town walk with mexican woman hold cardboard. police resist riots * no matter times what i tried. but video keep showing filipino woman around him i can't see any of mexicans what the ...can anyone else know why?😅
how can get loot of gmail
Try Hispanic
How to do voice consistency?
Yo guys when I try to use flow it just says this although I am subscribed. When I click "Subscribe" it says I am alrady subscribed. Endless loop. Please help
does any know how to make camera keep follow a moving subject?
The camera smoothly dollies behind the man at a low angle, staying centered on his back as he walks forward
or you could just the camera follows the man as he walks
Try logging out and then logging back in. Also make sure to use chrome
Tracking shot
there ya go
thank you just like drone tracking shot
Tried it, still same. I purchased 2 days ago still the bug
Dude that sucks sorry. Can you try just canceling it and maybe resigning up? Honestly, I don’t even know if thats possible
Hi, I have the same problem. is there a fix?
bunch of contents flooding on the internet

Hey you know how to make selfi shot ?
Where is the channel is it on Instagram or tiktok?
So I used musicfx to create an ambient backing track and then bought capcut. So 9:16 format.
format.https://www.tiktok.com/t/ZTjtxsDqL/
I tried a lot genarate but video dont have audio
I use model as veo 3 (text to video ) beta audio
And this is my prompt
Ultra-realistic emotional scene set in a typical Filipino provincial bus terminal during the daytime. A colorful provincial bus is parked in a busy terminal — tricycles, vendors, palm trees, luggage, and soft sunlight in the background. Inside the bus, a Filipino mother sits beside the window with her two sons.
On her lap is a 2-year-old boy who looks Korean — pale skin, soft round face, black hair with bangs. He is holding a small toy. Beside them is his 4-year-old brother, average build, not chubby or skinny, sitting calmly and looking outside.
Outside the window, the Filipino father stands filming them using his smartphone, his face full of quiet strength and emotion. The atmosphere is warm, nostalgic, and heavy with unspoken love.
Include cinematic lighting, shallow depth of field, and natural color grading.
🎧 Include voice lines in the output (as dialogue audio):
2-year-old toddler (Caleb, soft baby voice): “Daddy… I miss you.”
Father (Filipino male voice, deep and emotional): “Caleb, Kalex, mga anak… tiis lang kayo. Aahon din tayo sa hirap.”
can i create a video here by sending prompt
Do u guys write what dialogue to say explicitly in the prompt or do you just let VEO control that
I tell gpt
That looks really good, nice work
I was messing around with my prompt. I only have the google ai pro sub. so i used the veo 3 (FAST)
Lost my 600 credits, no sound changed prompts then sound but different language even though in prompts it clear saying in english , i want refund my credits 😦
Damn thats good result for veo 3 fast, can u share ur prompts with me
thats because some mexicans genuinely do look very similar to Philippinos,
they honestly look perfect to me, thats funny
is there a setting for the video to not show veo watermark?
du brauchst vor allem eine US-Zahlungsmethode, US-Gmail kannst du selbst easy einrichten, wenn du bisschen Google bemühst
geht nicht bei mir brauche us telefonnummer sonst wird der acc nicht zugelassen und alle nummern von 5sim und co sind gesperrt
Tello scheint zu funktionieren
ist das eine esim ?
I can’t download video.
It shows me like this. But when I press the download button. It stand still. Doesn’t happen anything. Does anyone know how to fix that?
Nice. What’s prompt?
I believe you'll need to subscribe for that.
We have this tool that can create absolutely anything and everyone does Bigfoot videos for clicks 🤦🏻♂️
LOL I feel ya
600.. more like 6000 here
Damn they need to fix this issue, free money for them not for us
probably blew 1200 just accidentally hitting enter 😛
the only fix is the ai learning
which means us doing
Hahaha worst thing is when u not adding scene in jump and u accidentally creating seperate scene
i mean we're early adopters of 2 beta products
Yes i understand but as long as they should help us too for refund some points
I just create a new scene each time.. I gave up with jump or extend
try musicfx too..
it's pretty cool for ambient background sound
I asked chatgpt to analyze trending tiktok memes and write a veo 3 prompt for a video that will most probably go viral..
when i can produce another video - I'm gonna be so rich!
This is what it spit back at me:
Create a cinematic, 8-second vertical video in ultra-realistic style. A serious-looking man in a business suit walks slowly into a quiet, elegant art gallery. As classical music softly plays, he stares at a large, dramatic oil painting. Suddenly, the painting begins to move: it’s a looping animation of a cat in a tuxedo breakdancing in space. The man nods solemnly, wipes a single tear, and whispers, 'Profound.' The classical music abruptly cuts to bass-heavy electronic dance music for the final 2 seconds. High-contrast lighting, no captions, perfect lip sync, cinematic camera movement: slow dolly in, then fast whip-pan at the beat drop. Use viral pacing and meme-worthy timing.
I still miss using Veo. So annoying
is ultra-realistic really a thing. I mean, what is more real than real?
I started posting videos on tiktok 3 days ago. I had 8 followers. Currently up to 25! let's see if i can hit 1000 followers by friday by making these videos
I got a few ideas.. hopefully veo 3 cooperates
what is that mean? i already up;paded first frame and gettin this when uploading the last one
anything wrong with this pic?
hey i have an active Google AI Pro subscription (Google One), but when i go to flow, it says i can not use it because i have not active subscription. can anyone help me out?
did you solve it? i have the same issue
Its for ultra
Flow can be accessed if yoy have a actuve Google AI Ultra,not a AI pro one
Do you still need to purchase tokens with Google AI Ultra or do you have unlimited access to create and edit as many videos as you like?
nah bro, you should get 1000 credits
When will Flow be accesible in Europe/NL ?
Same here. I do have AI Pro subscription. But it says I have 0 credits.
What about with Ai Ultra ?
im aiming to get to 1k followers too with some new posts, good luck!
if any of you guys follow me on tiktok or insta let me know and i'll follow back. Happy to see what you make
Google really need to fix outputs that show game graphics even when the prompt is clear that it wants photorealistic real life footage! I've wasted thousands of credits because Flow decides to give me game graphics I didn't ask for! If anyone knows a fix to this, please let me know!
kinda annoyed when generate game graphic and cartoon video..
same any fix?
hey Brother. yeah man of course, here's the Complete Prompt.
(Subject: Dr. Aris Thorne, a woman in her late 30s with sharp, intelligent eyes, dark auburn hair tied in a messy bun with a single silver pen sticking out of it. She wears round, thin-rimmed glasses, a grey wool turtleneck sweater, and dark trousers.) (Action: She is standing on a rolling ladder, her hand trembling slightly as she gently pulls a colossal, ancient book bound in dark, star-patterned leather from a high shelf.) (Context: The scene is inside a vast, circular library with towering, floor-to-ceiling bookshelves. Dust motes dance in a single, dramatic shaft of moonlight breaking through a high arched window.) (Style: low-angle shot, cinematic, moody chiaroscuro lighting, emphasizing the contrast between the dark shelves and the bright moonlight. The camera performs a slow, almost imperceptible dolly in, pushing closer to her face, which shows a mix of awe and trepidation. Photorealistic, shot on 35mm film.) (Audio: The only sound is the subtle creak of the old ladder and the dry, scraping sound of the heavy book being moved. A low, tense, ambient musical drone begins to swell very quietly in the background. Dr. Aris Thorne whispers: "It can't be."). no subtitles, no on-screen text, no words
does someone know why i keep getting this? i am a subscriber but somehow i am unable to use gemini flow.
I asked support
HELLO
and they said it takes 24h to update
VEO FAM
they are lying
💀☠️
you won't be able to use it,
Anyone know why I cant sign up for a free trial for Google ai PRO? has the promotion ended
Anyone with the problem where they cant upgrade vom PRO to ULTRA? When I click the upgrade button it just shows a Text saying that I already own Google ONe
Has anyone successfully figured out how to avoid subtitles? I've tried negative prompts like "No on-screen text, No Subtitles. No Captions" but occasionally they pop up, and make no sense.
...then there's audio, still no dialogue generation with frame to video.
Yeah it's def a bug right now and there's no clear cut way to avoid them.
there's like 12k credits
Does anybody know how to contact Google Labs regarding accessing VEO3 as a creator?
What are the prompts?
Prompts are in my server from where the videos are forwarded
so when i try to extend a video onto another it says its not compatible
Any update about veo 3 google flow launch in india ???
bro
does someone know why i cant seem to access flow althgough I have the right subscription?
Anyone has a solution on trying to extend videos on flow with audio?
May be your country is not eligible for flow
Only Uk has access in Europe
i am in the US ;-;
Oh then you should get access
Today I bought the VO 3
Contact to support team @rain crow
my account had a german billing adress still...i guess that messed it up
and, It gave me 1000 credits
Guys VO 3 are not creating the faces of famous ppl.. what should I do now
nothing haha you cant do anything
i mean what you wanna do
VO 3 should be able to make faces of famous persons like trump, etc
i think people have created trump before, i have seen videos of it.
But it didn't work for me
Is Sundar Pichai in this server??
😭
lmaooo haha nope
sad
if you want to generate famous ppl maybe use custom workflows locally
how
whisk?
or to upload the pictures
no like on your PC
@main saddle plz turn off the slow mode, we have only 2 members in the chat
i dont think they appreciate you tagging the main team like that ;-;
what U mean like using Photoshop?
or any other software
no my guy i mean comfyUI
Hello guys! Someone is having trouble when trying to download 1080p clips from flow ?
I downloaded max 720 p,, So IDK
Any tips on achieving voice/accent consistency?
Seems ok for me but can take a little while sometimes
Im trying and im getting this alert every try
its says that the video generation couldn't be done :(
Hello, I got the AI PRO version for free for 1 month on my Google account, but on Gemini with VEO3, I am restricted to 3 videos per day, I see that on Flow I can also create some but it requires credits, do you have the solution?
To anybody that’s monetized on TikTok, are you having trouble with your Ai videos getting banned for not being original?
Hello, I have made a few videos on Flow, but I am having trouble finding them within Flow, can anyone help me with looking into the history of my own videos on the sight? It seems to be lacking. Thanks!
i hate EU for this
Whats this issue
How do you monetized it on there ?
i Got the same issue. it states that i need credits, but i'm a subscriber and got 0 credits.
No, what kind of content are you releasing?
same. are you using vpn? what country is in your billing address?
Netherlands and vpn
think the issue is the location of your google account. try using an american google account via vpn (create a new one) you might need an US based credit card tho rewarble might work
Does anyone know if one can move a Flow clip from one to another project
What's new?
Pffff
hey, anyone had issues with generation in flow in the past hours? I burned like 1500 credits today bc of failed generations due to no reason or policy reason, also the last 4 videos had no sound at all
that's odd, they are supposed to return your credit if flow didn't generate your video, and usually if it is against their policy then it won't get generated.... though I admit, sometimes the AI will generate it, but either it won't follow your prompt or it will follow it but do it quickly or out of frame.
When does flow release in Denmark??
they should have a recycle option, like if a clip was never downloaded and you try to recycle it, the clip will be deleted and they'll return maybe at least half (maybe too optimistic) of the AI credit, or even just portion of it.
Hey guys how do I extend a scene so it’s the same character I generated just saying or doing something else
i have had zero luck with this
my last 5 generations have zero sound as well smh
yo everyone it keeps saying Google flow is not available in my country
Im in the US
Anyone know how to get around: "Our policies prohibit uploading of prominent people at this time. Please try a different image or send feedback."
I'm using an ai-generated charachter that doesn't copy anyone's likeness.
anyone know why every single video i generate on Flow has absolutely no audio?????
Check your generation settings, make sure you're using Veo 3 - Fast or Quality. And only use Text to Video.
Yeah it's Veo3.. something has been up the last few hours
seems to get flaky at times. probably heavy use periods.
I've been having the same issue but when using a non human character. The audio is not working and I'm using Veo 3- ( I have tried both fast and quality)
someone make a water gun war video.. live from madrid.. tourists take the square!
don't make me bring yeti over there..
Hi guys, I am new, so forgive me if i am asking an obvious question:
At the moment, I am unable to further (extend or 'jump to') my videos with veo 3, only veo 2, is that normal?
Each time i try to, i get "Switching you to a compatible model for this feature. Submit again to confirm (or check settings for details)."
But veo 2 comes without sound.
The yeti says,"Shout! I mean shout! I said shout!" yeti shakes his head, "damn u!"
fine i'll do it.. 😛
Is there a work around?
sorry, yeah that's normal that functionality isn't there for veo 3 just yet
just re-use the prompt and create a new video
don't extend.. don't jump
always check the veo 3 setting before hitting enter
is anyone else having issues with Veo3 not outputting audio tonight? ive burned through like 8000 credits, and maybe had 4 videos with audio. pulling my hair out!
I havent started yet, but there's been some comments
comments about audio not working?
yeah
ah ok, thats comforting, thanks!
I've noticed it gets flaky at times you'd expect heavy use.. think they run into some compute crunch at times
wild theres nowhere to submit a support request. id quite like a refund for those credits
yeah.. the compute time costs money whether it works or not LOL
fair
i guess 😛
probably should have given up for a bit
so much money..
got 13k followers on tiktok in 48hrs, so quite happy to blast through credits to make more content, as now im on the creator rewards program
one video just about to hit a mil views
nah they only pay you a) after youre in the program and b) for videos that are 1min+
id prefer not to share publically!
9 mill aint public?
so ive got a script written for a 1 min video, but i cant get any damn videos with audio
that's a lot of friends bro
9 mil?
its public, but its not sharing to a group of people who can replicate it 😉
nah i decided to steer away from those guys as seems to be a LOT of conten
content*
I've been aiming at 1 minute for my videos
good. There's soooo many. I got an idea for a new take on my yeti, but I'm trying to do some other stuff.
supposedly $0.40-$1.00 for every 1k qualifying views (watched more than 5 seconds, came from fyp etc etc). thats silly money if you get virality
Yeah, would be nice
say half of a 20mil video qualify, thats $4k-$10k. madness
ill believe it when i see it
yeah if you can do a few a week that's pretty nice 😛
if you find a good voice dubber that can dub a consistant voice and has regional accents let me know please
eleven labs...
i couldn't find accents
search for the accent you need
oh wait that was capcut
hmm think i looked there can you put a video in and have it do the voices?
on eleven labs, you just upload your video, then it spits out the voice in the same pacing, so you can overlay it and its perfectly dubbed
yeah its their tool called "Voice Changer". got to pay to use it, but its not much
struggling to get veo3 to be consistent with accents. pain in the ass!
but as of right now, im struggling to get it to give me any audio. so think im gonna call it a night
happy prompting!
yes audio is dead right now\
I lost tons of credits because of it
ive given up. painful
yeah im gonna try again tomorrow
yeah eleven labs does what i want.. just not very good lol.. the inflection is bad
for a comedian.. that's everything
yeah it can struggle to go from one specific accent to another. say canadian to american. IF veo3 spits out a canadian accent everytime, and you want it to be canadian... you could use elevenlabs to make the voice consistent
is flow only available int he united states now?
will it work with vpn if im int he UK?
I dont have the option to start project?
is it available in your area?
yes I can use it on my phone, but first time using on pc
hmm.. logout and back in.. clear cache all that
tried it, not sure why mine doesn't show me logged in like this.
on my phone It shows it
weird
post in flow bugs?
it got the last word wrong though.. i think.. i don't speak spanish.. LOL
anyone speak it?
chatgpt gave me this for the news banner
Batalla Acuática en la Plaza: Turistas Americanos Responden
does that say water gun war?
nop
LOL
i mean im spanish, i dont know the exact traduction in english
Water gun war update: Americans take plaza!
but maybe is like "water war"
that's what i want it to say
thats better
that could be
Batalla de pistolas de aguas: Los americanos conquistan la plaza
It really hates doing captions right
i never tried
the videos are funny though
best solution maybe could be add the caption manually
I dont know what you are generating but remember that childrens do not have voice so if a children is talking the whole video doesnt have audio
I wish I had ultra
every time i tried (veo3 - quality) option, quality is much worse than (veo3 - fast) option T_T
im having issues with audio as well.
how can i generate video for free?
you cant.
still?
Hola necesito ayuda alguien que hable español?
hey, I am new here. Hoe can i create a video?
I started a project in Flow and can't continue with the same setting. I really like the animation but it wont record audio anymore and I get an error.
how can i make flow generations 9:16 aspect ratio
I have the same error
You can run your output through Vegas 22 AI Auto Reframe.
you cant. have to use Runway another program to do so
bring your veo renders into runway and expand
Has anyone reached the end for their first Ultra monthly sub? Did your credits disappear or carry over? Did they give you another 12,500 credits? Probably too early for anyone to answer. I have another 8 days to go.
yo how come i dont have any audio??
no one does
anyone with the same problem?
veo wacked today
I just made this a few seconds ago. Veo3 fast.
I think it's specifically the non-human characters that are not having audio right now
Stupid question but where is my library? I lost that when imagefx was replaced by whisk
Rate my work please
its back!
is anyone using veo3 to make longer videos (i suppose making multiple 8 sec clips and putting together) ? is anyone monetising content created with flow?
would love to connect
Me too.. I'm trying 😛
it's insisting on putting captions in everything i do right now.. i've tried everything to tell it no captions.. no subtitles.. it just does it anyway
Any news on when possible be available in Sweden?
how do I fix the not available in country yet but im in America?
I can't get any audio in veo3 flow
Nice one,
Check the audio availability section here flow-announcements┇📣
The only times I didn't get audio it was either set to the wrong VEO model or the prompt referenced non adults (which are words even blocked from here, sheez). I can't even link to flow announcements which it totally weird.
seriously, what is its fascination with captions.. terrible
it's just too expensive to keep hammering out attempts when it just does random stuff that ruins it.
Took this prompt about 15 tries to produce what i wanted.. most were spoiled by mispelled captions that i didn't ask for.. I'm putting a background in during post. That way it's consistent.
15 and it's that simple
3 more in a row with captions
A realistic, no caption, no subtitle, cinematic 8-second shot: Ambient sounds of a comedy club. Sam is an 8 foot tall Yeti. His towering frame is massive yet dignified. He has thick, shaggy white fur, with a slightly darker grayish-blue face, giving him a rugged yet noble appearance. His deep-set, intelligent eyes glow faintly with an icy blue hue, and his brow is heavy but expressive, hinting at both wisdom and strength. He has a voice that is a deep baritone and he has a humorous, sarcastic delivery. The background is solid green. Close-up, The Yeti is center frame, he holds a microphone like a standup comedian and says, "Hear me out," speaking animatedly with subtle hand gestures after a pause he continues, "You see, - all my friends are felons! - yep, not a single voter among'em!"
i've tried various wording as well as just not having it
makes 0 difference
$30 worth of captions..
4
I'm from SEA region, how to by pass to use flow.
it show "Flow is not available in your country yet" in spite i'm using VPN at USA.
Think I'll have to wait for veo 4. veo 3 is a messy tease
alright, well of my $50 of credits.. that's 50 videos.. I got exactly 5 without captions and only 1 that was what i needed. I'm out of here. See ya next gen.
complicated stuff.. really
gonna cost me like $500 to tell a 1 minute joke
LOL see should be that easy
I just threw your prompt in my prompt generation app and that was the first attempt.
This was made with VEO 2 fast I think
try making a new project if you hadnt already, sometimes it gets weird in the same project w/ flow.
finally
same prompt
but it's just too random to be productive.. i mean if i had a bottomless wallet maybe
its like 1 in 10-15 to get a usable one
and it would probably take another 30 tires to get one that was more consistent with the first clip
and the only thing i change is the monologue
I guess just commission a yeti suit at this point.
LOL
Yeah i did that. didn't seem to make a difference. What app do you use for the prompt? I've been using chatgpt
I think the captions have to do with wpm.. i get that feeling
Created my own, finalizing things and then I'll release it soon.
what did you use to create this video
You know you can remove the captions using Capcut using the "AI Remove" feature and you can just outpaint it
Yeah I get it, but I don't think even Google knows why it's doing yet so for now, we'll have to work around it haha
man that worked great thanks
just bought capcut yesterday
those were driving me nuts.. LOL
woohoo 2 clips with the same monkey 😛
LOL
Today a good day for creator
U can use whisk to generate a picture of your monkey, then u can ask whisk to give a prompt how he looks and then tell gemini or chatgpt that this is allways the base for the looks, is more consistent. Maybe thats possible for the sound of the voice aswell? cuz this is my biggest problem that in each clip he can sound diffrent
voices are same for me generally
Mhh maybe cuz im letting my character speak german
How are some of you guys generate things so realistic? What I generate with veo3 is nothing close to this, are you using different prompts or something else is post-processing?
just add like POV Camera etc
35mm or selfie stick
i was saying photorealistic but it still looks cartoonish so I guess we need to specify the camera model etc
Could you share your prompt please?
check your dm
The greenish glow of the night vision flickers slightly as the camera, attached to a selfie stick held in his furry hand, moves with each cautious step.
Bigfoot looks at the camera with a worried expression and says nervously:
"Hey guys... I think I’m in big trouble."
The eerie glow of the night vision highlights the dark trees and underbrush, casting long shadows along the forest path.
try this guys and share 😄 im wondering 😄
how to make creature more photorealistic mines are cartoonish..credits are wasted
I suggest don't use " " or the words Said or Says in your prompts. I don't get any captions in Fast. Something like the below works for me.
eg, ..The Yeti is center frame, he is very animated and holds a microphone like a standup comedian, Hear me out he declares, and continues after a brief pause, You see - all my friends are felons! - yep, not a single voter among'em!
Can some one creat a video for me in veo 3
Guys, does anybody know why when I use scenebuilder and generate a second shot - it generates without audio. Is it how it's supposed to be?
I have the same, if u look top right it says failed 2 generate audio
any free ones you may know about?
Can’t get my talking alpaca to stop chewing grass or to lip sync. Had gotten it perfect a few times and passable a few times over the weekend, but last night nothing would work at all
no, there was no message like that
Hello, I apologize for bothering, but is there a way for me to access Flow or Veo 3 while I am in Kuwait? cause it doesn't seem like it's available!!
Yes use vpn
I am in Saudi using veo3
Comrades, there used to be a place in Flow where you could loop the videos you made.
Where can I access it?
anyone can help?
I'll give it a try, thanks
I'll try that too! thanks
Nope, didn't work.
.
Left some words in my prompt that might have confused it.. still waking up..
next one looked right
working much better thanks
Trying whisk. I used chatgpt to create the character prompt. I generally get a couple variations, but it's the voice that changes most
Yea i struggle with constant voices aswell, prolly need to tell chat gpt a base prompt for that aswell
Would be nice to have a good lyp sinc in veo3 but i guess the workaround is still using labs
Cant be true, i can def set up the mood or if they are screaming etc
yeah that's what i noticed
But u can use the same prompt 4 times and everytime it sounds diff
that's more mood than voice though
Yeaa ig, but pich changes with mood etc so it has some control
U work with elevenlabs?
yeah, it's not right
U can basically read ur text in, like with all the pauses etc u want and just swap out the voice
for jokes try "in a comedian style voice sayuing this joke"
works for me
in the prompt or elevenlabs?
yeah sometimes it's really good
I'll adjust my prompt with that and see what happens
why does flow say not available in my country when i logged in yesterday and im in the US
are you a democrat or republican? LOL jk 🤣
sorry, just read an article about some singer saying if you're republican our shows aren't for you.. cracked me up
?
joke
so how do i fix
Hi, sorry as im not familiar with flow.. and discord 😅 anyway.. how do i download from my library to my iphone.. it shows download icon but when i click nothing happen.. is it normal?
Hi. If I take some pictures of a setting, can I have the AI build the video with the settings background? And perhaps a picture of the person that I want in the a I video
For example upload some photos of a corner store with shop keeper then use Ai to describe a scene in that setting. All of my videos have no sound. How do I solve that as well
And what's the best way to use speech? Put it in quotation marks
Il y a des français qui ont accès à veo 3 avec + que 3 vidéos/jours ?
Heya guys, I'm pretty new to VEO and Flow but not AI video and other AI prompts, however I'm noticing there's some things that veo just can't seem to get right. One of them is a detailed car crash from the interior, FPV. Any interesting prompt suggestions to make it realistic instaed of looking like a GTA crash? :)
Flow isn't available in the United States?
Hey guys, Im trying to access to Flow, however this annoying location country message keeps popping on, VPN is not fixing my issue, could someone help me with this please?
Hey guys how can i use hindi language in veo 3 ? It says its not supported
Currently Flow only supports English prompt
So you can't use other languages
How come i can see other people use on youtube or tiktok
you can't
prompt it in english and then for the audio parts where they say something put it in hindi
everytime i’m trying to create a video it doesn’t give me the option i got the Google Ai pro
if you buy the $20 subscription you're still limited to 10 8 second videos or can you pay for extra??
I hear they just put out an update that makes it possible to get alot more expressive
It doesn't work great on phones. Use chrome or edge not safari put it in desktop mode but even then ...
what about ipad?
don't have an ipad so not sure.
would expect you will still need to request the desktop version
okayy thanks
you need a laptop/computer. you can still access it with the google app instead of your safari browser.
How do I connect the 8second clips?
what country flow is available in?
I don’t use on iPad right
is the audio for veo3 working today? I generated one clip and it didn't have audio. i can't afford to lose more tokens, wasted a lot yesterday without realizing the issue was happening to everyone
Hello, I am in canada but when i try to access to flow https://labs.google/fx/tools/flow it gives me this error: Flow is not available in your country yet. But I confirmed that Flow is indeed available i canada, can anyone help?
Google Lab's AI video generation tool, powered by Google DeepMind’s latest model: Veo 2
I'm the US, why does it say it's not available in my country?
either a video editor or scene builder.
How do I make it so flow veo 3 only makes 1 vid at a time instead of 2 which uses 200 credits
Wasted a ton without realising bit sneaky that
Nvm it’s literally right there I looked over it ahaha
Me: Hmm, why is my sasquatch snowboarding up the hill?
check prompt: is snowboarding at high speed down a steep snowy trail
Me: obviously that's why.
lol make it a video of him taking a dump and laughing at the camera
is buying credits not available on AI pro plan?
You can upgrade to ultra...
Also, if you are just doing 8 second clips and stitching them together in an editor you can use gemini
It shows flow is not available in my country
It would be wrong to tell you that a vpn can bypass country level content restrictions
I mean, ik that but it takes approx how much time to release it everywhere
They won't release it anywhere hostile that might try to say sue them over it.
Nice point
yooo
How come you have to pay for failed generations that’s kinda bs
If it fails it should credit back. If it succeeded and you don't like it... 🤷♀️
anyone knows why my videos not getting sounds. I am using the veo 3 highest quality 100 creds per video.
why they go uphill.. beyond me
Hello, how can i fix this prob?
is there a way to replicate a video but have the faces of the people in the video change like if i wanted to animate an already made video
Breaking News: Loch Ness Monster finally found! Lol.
How can I preserve the face of my character created in flow /veo3. In the same project, vary the character in other scenarios, so as not to hallucinate and change the created avatar. I'm having this problem
Is my best option for Flow to upgrade to ultra? Im on the monthly AI pro plan rn and I was just testing to see how everything works and im already out of credits. I wanna have high content output but seems like the yearly Ultra option is the only way to do that, just wanna check if im right?
Yes you'll need ultra. Gives you the most credits and allows you to buy more as needed.
Feedback?
Alr thanks appreciate it
is veo 3 broken right now? its giving results that are totally mangled
Somethings def off today with it
veo 2 fast is making stuff that looks better! wth
Yep I shouldn't have used my credits today 😭😭😭
they gotta refund for this, this is not cool
Has anyone done a shortfilm in Veo?
no audio for me either
anyone know where i can learn how to make better prompts
the things that its outputting arent as vivid and clear as id like them to be
also is there anyway to get rid of morphing and unnatural movements
like the fingers and other limbs are merging and dividing and doing all sorts of weird stuff on my character
that is so cool wth
how
Hello, I'm generating videos with Flow, but I keep getting videos without audio, just captions
okay, i'll try veo 3. Do we get credit for the failed generations?
It credits for generations that fail to generate, not ones you generate using the wrong model.
no problem. What if it generates with incorrect audio?
can someone let us know if veo3 is working for them now with audio
Also, I'm not getting consistent character generation.
when I'm trying to generate video with consistent character using "Extend or Jump to" function, it automatically switching to Veo 2 (without adio)
Now I just generated with Veo 3 fast (audio beta) and the audio is gone. This is too inconsistent.
i'm also getting "audio generation failed" and i'm trying to use 'image to video'. got a few to work, but it mostly fails and adds subtitles in lower third
What is your prompt?
Character varied slightly as I went along. I guess I need a more descriptive prompt.
i got the audio to work on one. here's that video, but the dialog is gibberish. i've seen other users say they're seeing the same issue.
whats the prompt
the prompt was super simple....standing man speaks "Man... this thing is taking on a life of its own."
any tools anyone is using to generate and add audio to veo2 generated videos?
[SAUCE ALERT]
#################################
Honestly Claude ai does a good job, also... I've found that saying (write the prompt in the style of a clip text encode or a clip vision) does well.. I'd also make sure to tell claude: "Hey do this in the first 2 seconds so that i have time to do this for the next 3 seconds, and then end that so i can do this for the last 3 seconds"....Is a good way to fit things in one vid.
#################################
I think just give it more context and instrruction and youll be set
use an ai or something
guys flow is avaialble with Google Ai PRO or is a higher level?
Is there a trick to getting audio working for these? I'm trying to animate a rabbit talking and it keeps making strange sounds...not speaking...hehe
yeah I've had gibberish on all my videos, but I'm trying to have a rabbit talk...
guys flow is avaialble with Google Ai PRO or is a higher level? someone can response me please, thanks
Yes pro gives you limited credits and 3 free generations through gemini / day
Im figuring out how to get a initial video in Flow, but i am failing to have continuing video with the same image, and extending the scene. its wanting to switch me to Veo 2 Fast which doesnt include audio, so i keep getting the 2nd scene with no audio. any suggestions?
this dont make any sense, I'm trying to have audio and I dont kno how to add it
you have to make sure you have Veo 3 - Flash selected, Veo 2 does not include audio
guys, the ingredient to video doesnt work on veo3?
i think its because veo 3 is limited, ingredients and extend scenes are veo 2 features that do not include audio
but im not 100% sure, completely new to all this
ahhh that makes sense
Same is happening with me
The only way I've gotten it to work is the first 8 secs with veo 3 fast... then the rest with veo 3 quality. And it. Maintains the same characteristics and stuff
and that uses a lot more credits, but if the quality is good its worth it?
is veo 3 still giving mangled outputs?
Probably my favorite thing I’ve made so far
U made the dialogue your self? Pretty funny 😭
Sometimes you just need to describe your character as consistent as possible, like their physical descriptions and clothing, and Veo3 has a better chance of repeating them, even without extending them, like this.
and this youtube example, you could tell the guy's face is changing face a lil bit, but it is close enough: https://www.youtube.com/watch?v=daVuK-3ypl8&t=2s
This is the recovered datalog of a Jedi Knight vlogging his day on Coruscant, just as Chancellor Palpatine initiated Order 66.
What began as a normal day, fixing a faulty lightsaber with a new kyber crystal, visiting the Senate, and grabbing a bite at Dex's Dine, quickly spirals into a desperate fight for survival. This is the last transmission...
that is composed of 3 (8 secs) clips btw, i just copy and pasted the description of the main guy on my 3 clips, with 3 different action/scenario and hope that veo will generate him consistently, lol
So how many of you have completely aboandoned Veo3 Quality and just go with Fast
never even t ried fast since quality was so bad LOL
Quality is bad? Compared to what?
compared to my prompt 😛
haha
Couldn't get my sasquatch to snowboard downhill all afternoon.. every clip.. uphill.. I might make a video called magic mountain.. "look, we're going up!"
what was your prompt? maybe you weren't detailed enough? Veo be really clueless alot of times
What happens if I use fake vpn and edu to access google flow?
i mean, when you put in downhill and they go uphill..
i tried, down the mountain, downhill and just nothin about direction
i have no more money for credits so.. i'm off to sell some plasma.. back in a few!
A playful and whimsical video clip opens with a wide-angle shot of a snow-covered mountain, bathed in the warm glow of a late afternoon sun. A large, male Sasquatch with light brown fur is seen "snowboarding" uphill with surprising ease, his board leaving a smooth, continuous trail in the fresh snow. The camera, from a low angle, tracks alongside him, emphasizing his effortless ascent. The setting sun casts long, soft shadows, creating a magical, dreamlike atmosphere. A subtle motion blur on the snow further highlights the uphill movement. The sound of gentle whooshing accompanies his progress, punctuated by the faint, muffled sound of a happy male voice exclaiming, "Look, we're going up!"
i had a few where they were going uphill and sideways
wait you put in uphill and it went downhill..
LOL
let me try generate it again
Its June 18th here
i notice they gave same kind of voice to the sasquatch
even mine also sounds like yours LOL
still at 0.......
sometimes its just a trick on the eyes.. he might be going down from that angle.. but why choose that angle. I've been trying to get to do a chase angle, but they're all from the front..
hmm that looks good for fast
i know right
idk why quality mode has been miss for me today
i've generated like 4 video with veo 3 quality and all of em were bad output
pretty sure it has to do with the load on the servers
wish they had a little guage or something that would tell us peak and non-peak hours
A dynamic, wide-angle shot follows a male Sasquatch, characterized by long, shaggy brown fur and a muscular build, as he carves graceful turns downhill on a snowboard. The camera is positioned behind him, slightly to his left, maintaining a smooth chase angle that emphasizes his downward momentum. Snow sprays up from the edges of his board with each turn, catching the late afternoon sun. The background consists of a snow-covered mountain slope with scattered pine trees, rendered in a realistic, high-definition style with natural lighting. The focus remains sharply on the Sasquatch, with a slight motion blur in the background to enhance the sense of speed.
oh wow lmao
yeah that's the shot i wanted.. nice
pretty much the same description as far as camera goes
yours is a bit more descriptive
I'll give it a try next time
i had to use prompt assistant for it 😅
same! , but nowadays i went back and forth with using gemini and chatgpt
I just add to the character description to customize it a bit.. like the flags.
since it won't do accents
A dynamic, wide-angle shot follows a male Sasquatch, characterized by long, shaggy brown fur and a muscular build, as he carves graceful turns downhill on a snowboard. The Sasquatch is wearing goggles on his forehead and a green, white, and Mexico flag draped over his shoulders, flowing behind him in the wind. The camera is positioned behind him, slightly to his left, maintaining a smooth chase angle that emphasizes his downward momentum. Snow sprays up from the edges of his board with each turn, catching the late afternoon sun. The background consists of a snow-covered mountain slope with scattered pine trees, rendered in a realistic, high-definition style with natural lighting. The focus remains sharply on the Sasquatch, with a slight motion blur in the background to enhance the sense of speed.
somehow it doesnt want to position the camera from behind
can they talk with fast?
yep!
yeah it seems to try hard not to
doesn't like going downhill either
try getting it to do tricks.. i had a hard time with it.. it would stop and hop and not do at all what i described
i was showing a friend today. And he's like just do what this guys doing.. and he shows me the bigfoot vlog.. and I'm pointing out all the flaws.. like, it's a different one every time. it has captions. they're all misspelled. this guy doesn't care about quality at all.. he's like.. yeah, but it's funny..
its SO frustrating when you are punished for bad model output
I wish you could submit good feedback to google and get your credit back
I've been suggesting that, like a recycle option, as long you didn't download the clip, you can "recycle" it, and even just get half of the credits back.
you'd think a multibillion dollar corporation could foot a few credits
recycle, like it will be deleted and return like half of the credit
I've a pro account with 1000 credits that I don't use any one want it ?
lucky lol
even if you did, if you provide detailed feedback google can learn from you get it back
oh yeah
seems like a good way to get high quality data on where the model goes wrong with detailed feedback
another prompt that is hard to make is someone winning an arm-wrestling match, not sure why Veo won't allow it, I tried Veo 2 and 3, and always failing... anyway, a challenge to anyone. I got this prompt from chatgpt and it doesn't work, even just basic simple prompt like guy1 pinned down guy2 in an arm-wrestling match, wont work
Two average-built men are caught in the final moment of an intense arm-wrestling match. The man on the left, wearing a blue shirt, is decisively winning—his arm locked in a powerful angle, forearm tight and vertical, with his wrist slightly flexed inward for control. His hand is pressing the right arm of his opponent down with firm, steady pressure. The man on the right, wearing a black shirt, is clearly defeated—his wrist is fully bent back, knuckles facing upward, and the back of his hand is touching the tabletop. His fingers are splayed slightly, showing loss of grip and strength. The blue-shirted man's fingers remain tightly clenched around the other's hand, guiding it firmly into the pin. Their elbows are grounded, muscles taut, and veins pronounced. The table shows slight vibration from the force. Style: cinematic realism, medium close-up, sharp detail on hands, tension, and the moment of final contact.
i give up
A dynamic, wide-angle shot follows a male Sasquatch as he carves graceful turns downhill on a snowboard, snow spraying up from the edges of his board with each turn, catching the late afternoon sun. The 3rd person view camera is positioned behind him, maintaining a chase angle that emphasizes his downward momentum. The Sasquatch, characterized by long, shaggy brown fur and a muscular build, wears goggles on his forehead and a green, white, and red flag draped over his shoulders, flowing behind him in the wind. He is expertly navigating a pristine, snow-covered mountain slope, surrounded by towering, snow-laden pine trees under a clear, bright sky. The focus remains sharply on the Sasquatch, with a slight motion blur in the background to enhance the sense of speed, creating a realistic, cinematic feel. The ambiance is crisp and cool, filled with bright, natural light. As he glides, the distinct crunch of snow under his board is heard, accompanied by the whoosh of wind past his ears. He lets out a boisterous laugh and exclaims, "They said I couldn't learn to snowboard. They said I was too hairy. Who's laughing now, gravity?!"
no matter how many times i refined the prompt , it just doesnt want to position the camera from behind. i dont even know where it get the idea of "sasquatch holding the gopro"
i tried, i used prompt assistant and i didnt even realize it modified the core idea of your prompt- this was veo 3 but apparently it failed to generate audio lmao
thanks for trying.... compared to snowboarding, I kinda assumed/imagined that this will be too easy for the AI, lol
gave it another shot , and yeah i give up once again
that was funny, they just smashed each other's hands, lol
both are using... both arms or somethin i just noticed that
sure, but we're all just throwing darts
lot of random to this thing
first video i generated, was wanting to create a fragrance ad