Building text to ASCII diffusion model, need advice and guidance

i wanna build a text diffusion model which interpret text and convert it into ascii images

so like

Text: build a cat

Output:

(o.o)

\> \^ So, i have a decent background of ml algo (completed cs229, cs230, Ml architecture and basic CNN and diffusion model)

ik making a project like this is tricky and making diffusion model like that from scratch is hard but i wanna try it because that's wot make me excited lol ...

I am currently reading GANs research paper, can u guys help me in finding more papers which helps me in making this project or guide me through this good title for this

Thx in adv