Hi Vijayta30,
Thank you for sharing your experience with us.
Text generation within AI models is a known limitation, and Firefly is no exception. When generating images or videos, the model treats text more like a visual element rather than structured language, which means it can sometimes produce incorrect, inconsistent, or even different languages, especially in stylized or complex scenes. Even when prompts clearly specify English, the model may not reliably enforce it.
One good option is to generate the image or video without the text/symbols and then add them afterward using Adobe Express, where you have full control over the language and accuracy.
Unfortunately, credits are the cost of each generation, so they can’t be refunded for unwanted outcomes.
Hajar
[Moved to questions]
Sign up
Already have an account? Login
To post, reply, or follow discussions, please sign in with your Adobe ID.
Sign inSign in to Adobe Community
To post, reply, or follow discussions, please sign in with your Adobe ID.
Sign inEnter your E-mail address. We'll send you an e-mail with instructions to reset your password.
