https://www.reddit.com/r/PowerMetal/comments/1vrsvp3/a_trait_ive_noticed_in_power_metal/
A moderator closed my thread on there on the grounds of "posting up AI content". Please visit that thread and tell me where i actually did that?. Or does merely simply mentioning AI usage in a thread count as "posting up AI content" ?
I wanted to make a song earlier in the week that involved a few piano melodies from Magix Music Maker, but when i tried to, it told me the music was copyright owned, so therefore it forbidden the upload. Blimey, how strict is the filter on Suno. As far as i know, all musical events and packs from Magix Music Maker are run by Chatoo or Cathoo (forget its spelling). I know that MMM events are not royalty free, not if you want to use them commercially. Obviously the same thing would happen if you tried to upload a song that you made using MMM, and you used only the 1 of its musical events in your song
It doesnt matter whether you are from the USA or the UK, i think both countries will have the very odd business's out there that will post things like this in their shop window or doorway,
"No racists
No homophobes
No transphobes
No bigots
Are welcomed into this restaurant"
Now here's the thing........you can bet your bottom dollar that ironically those restaurant/cafe owners are going to be among the most judgemental people you will ever meet. I mean why even post up a sign like that???, how are you going to know if any given customer that walks through your door is or isnt any of those things?, inless they come in and actively do any of those things, or you eves drop on a conversation their having on their phone or with the person they came with. I reckon a sign like this is just a dog whistle. The message that they are really trying to put across is "come in here if you are on board with woke beliefs, go away if you are not".
Because how many non lefties out there are actually ever any of those things??, the truth is you would be hard pressed to find any that are. The vast majority of the times when the far left and woke call you those names, they are just nothing more than begging the question fallacies. And some of those things that far lefties and wokes think is transphobia is not transphobia, its just someone with fair beliefs. I actually think its a more fairer stance to want to protect the freedoms that cis women actually fought for for years, like a bathroom of their own, and sports divisions of their own, than support a demographic that wants to take liberties on those rights.
I dont know if every AI generation site is like this, but on some you can create AI characters, this helps alot with maintaining visual consistency. Im using Mage Space where you can create characters on there, but there is one factor that is missing which i think is important............height. Why is there no where where you can select the height of your character?. It seems to just assume that your male is gonna be a bit taller than your female. Well some women out there are taller than most men, ect ect.
Obviously there is no need to type in size and weight, as that can be seen by their profile pics, but height on the other hand is a different story. You cant tell how tall someone is just be looking at a photo of them where they are the only person in shot (inless their stood near a door or something).
Not that long ago i made a thread on here asking what good free AI voice generators there were, one that would do emotions, and not just say the words clinically. One place i was recommended was Scenema, and yeah it seems great for emotional dialogue, but although they aint strictly speaking free, their free tier gives you 500 gems per month, which is plenty, if all you wanna use it for is voice generation.
I still have 472 gems available, but now i can no longer use the audio only section :-( . I get a message telling me that i have used all of my allowance for the month......what?????.....ive only used up 28, how is this so????.
I know of Kororo and i know of the guy on here who created his own free text to speech site using the Kororo models. I also know of this place https://ttsmaker.com/ , that allows you 1K's worth of characters per week for free. I also know of Eleven Labs (but they aint free).
The trouble is tho, they tend to be emotionless and clinical. If you use a video generator that allows audio production aswell, you can write in the prompt how you want the person to say those words, you cant do that on a text to speech site, you can only type in the actual words you want them to say. What there really needs to be (if it doesnt exist already) , is a site that gives to an emotion option. So you pick the person first, then you select an emotion, like clinical, authoritative, joyfull, upset, seductive, sarcastic, playful, angry ect ect.
I took out the basic plan on Mage.space yesterday that costs $10 for a month, that one gives to unlimited access to its basic models. But i have to ask this question, were those models thrown into the reject bin or something then taken back out again?, cause they are so fricken useless and incompitent!!!!.
Today ive been trying to edit an image using these following models
Flux Krea, Flux 2 Dev, Pony Realism V2, Flux Schnell, Flux Dev and Stable Diffusion 3.5 Medium.
All but one of those models are more useless and incompetent than the models that are used on Video.Me and NSFW Lover, and that's saying something!!!!. I cant upload any of the images i created today as they are all NSFW certificate. But with all of these models but 1, they couldent even change the wall and the floor to the colours and material i wanted. They would totally change the appearance of the person already in the image, and even add clothes to them that i never asked it to. The only one that was somewhat decent was Flux 2 Dev, the others are more useless than a bent nail.
Now as far as settings go with these models, is it best to turn Steps to the slowest speed, Guidance Scale to the highest setting and Prompt Strength to the highest setting?, as thats what ive been doing.
Ive tried having a look, and when it comes to making AI generated videos, models on offer to use are things like Seedance, Sora 2, Veo 3, but do any of them produce audio for your videos aswell?, like make the people in the video actually talk?, i tried one of them and all it played was music over it, no actual voice from the man. Some of them needs credits before it will produce any video, I dont wanna waste money on buying credits if none of them will produce actual voices.
When i go into the models options, all im seeing is aspect ratio, resolution and duration, but no option for audio
Ive tried a fair few and they are pretty dumb to be honest, so many times they cant follow simple instructions. But yet the SFW only image editors like Chatgpt, Gemini and Dola, are pretty smart and usually do what you tell them to do (especially Chatgpt). Ive used Video Me and the FLUX.2 Klein model on NSFW Lover so far, and to be honest they are rather dumb, the latter often over saturises the pics, then i have to open it up in Photoshop and turn the saturation down.
I know that to find a smart NSFW AI photo editor, i will likely have to pay for it. But if there are no good ones with a "free tier" out there, then can you reccomend me one that doesnt charge an arm and a leg?.
Is there an AI image generator that works like how Video Me does, but is better?. Video Me will produce NSFW images but to be brutally honest, its actual AI is pretty thick, and it makes a pigs ear out of the most basic instructions and descriptions. Honestly, Dola is much more intelligent than Video Me but of course will not produce NSFW content. If there was one that worked like Video Me but had Dola's intelligence, that would be brilliant. I got a feeling that Video Me is chinese based because when i tried putting speech bubbles into an image, it made all of the text come out in chinese.