Link in Zwischenablage kopieren
Kopiert
I am trying to generate a logo with a desired text. It is not accepting it.
Text and symbol generation in images is not supported yet in the Text to Image module. The quality of our models will improve with time, and generated content will better respect the prompt. If you are trying to create just text, please try the Text Effects module. (Merging all similar threads here so we can track the number of reports for this)
@Anonymous37037930cx5u, While it has gotten better, Firefly (and most other AI image generators) is not good at putting exact text from a prompt into an image. Sometimes it is perfect, but usually, it does not get it right. The best solution is to use Adobe Express or Photoshop to add text in. The more complicated the angle or perspective of the text, the more likely you will need Photoshop to get what you want.
Please see this article on why generating text in AI image generators is hard:
This is a current limitation please add your vote to this bug. In the meantime, check out Adobe Express.
Link in Zwischenablage kopieren
Kopiert
That is a current limitation of Adobe Firefly. However, you should check out Adobe Express. Specifically the logo templates.
Link in Zwischenablage kopieren
Kopiert
When I put in a phrase like "Going to the Store" the AI creates an image, often with words, and the words are nonsensical such as "Gooein tausteeren" No one is ever going to use an AI image with nonsensical gibberish words! You should have an option to not include words/letters in generation and make it so that if there is going to be words, they're actual words.
Link in Zwischenablage kopieren
Kopiert
@Ethereal_Serendipity5CCD That is a current limitation of most AIs. Here is a post about this limitation if you want to learn more. I recommend creating an idea in our Idea (feature request) here.
Link in Zwischenablage kopieren
Kopiert
I wanted to highlight that the crucial feature missing in Firefly is accurate text rendering within images. It doesn't necessarily have to have spelling capabilities; it just needs to accurately replicate the text or names when they are part of an image request. In the long run, I'll be sticking with the image generator that excels in this area. I consider this priority #1.
Link in Zwischenablage kopieren
Kopiert
@Lonnie33081757qq7s When you say "accurate text rendering," are you referring to this type of issue: https://community.adobe.com/t5/adobe-firefly-discussions/text-is-wrong-in-text-to-image/m-p/13924887
Link in Zwischenablage kopieren
Kopiert
Link in Zwischenablage kopieren
Kopiert
This is a known issue with AI:
https://community.adobe.com/t5/adobe-firefly-bugs/p-text-is-wrong-in-text-to-image/idc-p/14080489
Jane
Link in Zwischenablage kopieren
Kopiert
I can't seem to create text properly from Firefly (TXT to IMG), for example words on a T-shirt or a building representing a word, text written on a blackboard... Firefly produces the result but alters the text! Does anyone have any ideas?
Link in Zwischenablage kopieren
Kopiert
Try the Text Effects module
Generative AI models have trouble with spelling. And Firefly is no exception. Except for the text effects.
Link in Zwischenablage kopieren
Kopiert
I tell the AI generator to spell a word and it always spells it wrong. very frustrating
Link in Zwischenablage kopieren
Kopiert
Text is currently a limitation with all generative AIs.
Link in Zwischenablage kopieren
Kopiert
Look at subject
Link in Zwischenablage kopieren
Kopiert
Do you mean you type a text that needs to be in the image?
Link in Zwischenablage kopieren
Kopiert
All AI image generators seem to be quite poor at generating useable text in an image. You should probably open the image in Adobe Express to add the necessary text.
Link in Zwischenablage kopieren
Kopiert
Link in Zwischenablage kopieren
Kopiert
Text-to-image AIs are (still) somewhat bad at writing plain text. This is a current deficiency for most Generative AI models - including Firefly.
The best solution for now is to use e.g. Photoshop to add your desired text in.
Link in Zwischenablage kopieren
Kopiert
I'm curious, what was the prompt you were using? I think it says Nashville or something.
Link in Zwischenablage kopieren
Kopiert
Hi there!
We appreciate you writing in. You can also review this article for more info on this: https://adobe.ly/4ccCNTC
Let us know if you have any questions.
Thanks,
KR
Link in Zwischenablage kopieren
Kopiert
Link in Zwischenablage kopieren
Kopiert
Hello @catparkerphoto,
Thank you for your message. I am sorry you are having this problem.
While it has gotten better, Firefly (and most other AI image generators) is generally not good at putting exact text from a prompt into an image. Sometimes it is perfect, but usually, it does not get it right. The best solution is to use Adobe Express or Photoshop to add text in. The more complicated the angle or perspective of the text, the more likely you will need Photoshop to get what you want.
Please see this article on why generating text in AI image generators is hard:
If AI image generators are so smart, why do they struggle to write and count?
Thanks,
droopy
Link in Zwischenablage kopieren
Kopiert
What Droopy said is correct. There is a way to "brute force" by creating a structure reference with the word "consulting" spelled out and having Firefly create a logo incorporating that, but it can be hit and miss.
Link in Zwischenablage kopieren
Kopiert
Digito a palavra e sai em inglês
Link in Zwischenablage kopieren
Kopiert
@gabriel3864844024aa Firefly and other AI models cannot produce legible text at this time.
Link in Zwischenablage kopieren
Kopiert
Link in Zwischenablage kopieren
Kopiert
@Ignacio3864877064xb No generative AI model can produce legible text at this time.
Weitere Inspirationen, Events und Ressourcen finden Sie in der neuen Adobe Community
Jetzt ansehen