No watermark Ai video avatar, and just for 0$

Hello everyone this is Ainz and today I’m going to show you how to make a talking avatar like this one for just 0$.

You are probably wondering right now there’s 10000 videos about how to do it. What's new ? Well I’m not showing you that method where you use d-id to generate a video with the d-id watermark at the bottom, and limited minutes as well as limited speech minutes, well this is not that generic tutorial today we are using the models directly on free hardware given to us by some amazing people to create our video.

And first we need a hugging face account https://huggingface.co this is the website of the amazing people go ahead and make an account.

Then lets get the action started,

PICTURE:

This is the image i generated:

And this is the prompt: modern disney pixar style dua lipa close-up portrait, clear face, black eyes, facing camera

To generate yours let’s go to this hosted stable diffusion it is what i used aswell https://huggingface.co/spaces/camenduru/webui-docker but unlike what people would usually do i want you to download my avatar and go to img2img upload it there copy and paste the prompt as well oh and try to add your own touch to it, change the name maybe add few keywords but keep the main ones “modern disney pixar style, close-up portrait, clear face, facing camera” then keep rerolling until you get something similar to the one i showed, clear facing camera not blurry, max the Sampling Steps enable Restore faces and Tiling also for mine i used DPM++ 2M in the samling method. Please note I generated around 14 pictures before I got that one. Ook now that you got your picture let us move to the next step

The Script:

So for our script we are going to use chat gpt ofc but we are going with gpt4 not 3.5 and I know a lot of people don’t have access to it yet but don’t you worry we are going to use it directly on huggingface thanks to an amazing person who hosted it for free,

Before i give you the link please make note that there are people paying for those stuff so you can access it for free and have a look at the future, if you can donate to them then do it if you can’t then at least don’t abuse it and over use it.

Here’s the link: https://huggingface.co/spaces/yuntian-deng/ChatGPT4 But please not it takes a bit of time to generates, I just went with chatgpt3 🙂 And for the prompt I went with this: “Write me a storybook with a rhythm about Ainz, a little kid who loves technology and wants to be part of the people who are shaping it.”

And the text is:

Once upon a time, in a town not far away,
Lived a little boy named Ainz, who loved to play
With gadgets and gizmos, wires and screens,
He dreamed of being a tech whiz, just like the tech kings.

Ainz loved to tinker, to take things apart,
To figure out how they worked, to make them smart.
He spent hours on his computer, coding away,
Creating programs and apps, every single day.

But here’s what chatGPT4 wrote after waiting for so long of cours:

Once upon a time, in a land not far away,
A curious boy called Ainz thought of tech everyday.
He’d dream about devices and machines that were cool,
The kind that’d make his friends in school surely drool!

Ainz had a great dream, the best you can find,
A dream to make tech that’s one of a kind.
He wished to be part of a creative lot,
The people who shape tech, the future they’ve got.

Every morning, Ainz rose before the sun,
Coding in his room - oh, it was fun!
Learning to create apps and games to cheer,
His fingers danced with excitement, like a tech engineer!

Ainz’s dear mama said with a smile,
“Have some breakfast, son, just for a while.”
While munching on toast, with his online class in sight,
Ainz’s tunes on learning filled the room with light.

Inventors and heroes filled his notebook so fine,
From Ada Lovelace to Turing, they’re all in line.
And when Ainz had free time, oh, what he’d do,
Tinkering with circuits and robots, like R2-D2!

One day, at school, during his favorite class,
Mrs. Green, a wise teacher, said, while holding a glass,
“Pay attention, dear children, science is at play,
The laws of physics will guide us today!”

Ainz took notes and listened closely,
His heart aflutter, he learned so rosily.
Physics and coding, oh, such great connections,
The building blocks of all his tech inventions.

Then Ainz had an idea and in his bedroom, he sought,
To create a device that would connect every thought.
“The Ainz-link,” he called it, his pride turning bold,
A tool that shared ideas, new and old.

He told his mom, his friends, and his teacher,
All about his dream, a future so eager.
They cheered him on, both far and near,
Their voices of encouragement, for Ainz to hear.

One sunny day in May, a letter arrived,
From a tech contest, the stakes were quite high.
With fervor so fiery, Ainz couldn’t resist,
This chance for the Ainz-link, it must exist!

He worked day and night, his eyes on the goal,
Acing at school, while learning code protocol.
The town buzzed with excitement, waiting to see,
A little boy’s dream turns to reality.

The day of the contest, Ainz was both scared and brave,
His Ainz-link device, ready to unveil and amaze.
The judges examined, asked questions, and pondered,
Could Ainz’s invention be the tech dream he sponsored?

Finally, it was the moment of truth,
Ainz held his breath, hands shaking, but soothed.
One judge took a step, an envelope in hand,
The contents inside were something quite grand.

“Our winner,” she said, “smart, creative, and free,
Is Ainz and his Ainz-link, such brilliance we see!
The future of technology, it comes with great stride,
With children like Ainz joining the ride.”

Ainz smiled in bliss, his heart full of glee,
He knew his dream was meant to be.
The people who shape tech, that’s where he’ll be,
For Ainz, the little boy who dreamt with glee.

Now let us move to the next step is

The Speech:

And you guessed right we are going with another huggingface space made by another amazing person: https://huggingface.co/spaces/mrfakename/tts this one is simple select english paste text get the speech and it is fast, really fast so we are going to move to the next one and our main dish

The Video

Now that we have gotten our picture and our speech we are going to move to sadtalker, another huggingface space that will let us mix the image and the speech into a video https://huggingface.co/spaces/vinthony/SadTalker

This one takes a bit of time to run but worth the wait, upload your picture and speech, check all the settings then hit the button and wait for your result to finish creating, and here’s mine https://imgur.com/a/5OBHApH

And for people who likes anime here’s a special bonus: https://huggingface.co/spaces/Plachta/VITS-Umamusume-voice-synthesizer