Is there a tool (or a prompt) in that I can reproduce the same exact image, but with translated words (like, from English to French)?
Hey! You might be able to use a model that can detect and extract images like this one: Text Recognizer Gpu | Charlesfrye | AI model details and then once you have the text you can use an LLM to translate it (or just google translate). Your last step will be to combine the new and old images, which you could do with inpainting. Filling in the Blanks with Stable Diffusion Inpainting
I think I get what you’re trying to do here. But there are a few issues that might get in your way. First is the non-deterministic nature of LLM image generation, meaning you never get exactly the same thing twice. This has nothing to do with the prompt, but with how the models work using some randomized values that make deterministic results unlikely. The next issue is, translation, once you’ve translated the prompt to a different language there is usually some degree of meaning lost in translation. (There’s an old movie of the same name). But if you could get a deterministic model with perfect semantic translation, then you just might be able to do what I think you’re trying to do. Maybe someone will figure it out. Although I could be wrong, I’d love to hear what others think