Falha no entendimento do prompt

[LukDsgn]

Estou pedindo uma frase em português e ela está sendo gerada na imagem em inglês

[Dean_Utian]

Could you give more detail to what's you have prompted and the result? Do you wnat text to appear in the image?

[droopydog500]

Hello @LukDsgn,

Thank you for your message. I am sorry you are having this problem. 

The model itself works on English. So if it encounters a language other than English and it is one of the supported languages (which includes Portugese), the prompt is translated to English and then sent for processing. At that point in the process, it does not know that the text is text to be displayed rather than acted upon. There is no easy way to get around this problem. 

One option is to add text outside of Firefly using Photoshop or Adobe Express (including using text effects in Adobe Express).

Another option is to trick Firefly into thinking the prompt does not need to be translated. One way is to add punctuation in the middle of words which you can edit out of the image later, like a period ("."). For example, if you are trying to get a sign to say "flores para venda" adding that text as something like "flo.res para ven.da". Might work. What you are doing here is trying to prevent Firefly from saying to itself, "oh, this is Portugese, I must translate it."

Thanks,
    droopy