This content originally appeared on DEV Community and was authored by Aman Choudhary
This is a submission for the Google AI Studio Multimodal Challenge
What I Built
Demo
Multimodal features
ArtfulWhisper is fundamentally multimodal, creating a seamless flow between text and image data to deliver its unique functionality.
Text-to-Image Generation: The primary multimodal feature is taking a user’s text prompt and transforming it into a rich, complex image using the Imagen 3 model. This is the creative heart of the app.
2.Fusing Text within an Image:The application then takes a second text input (the secret message) and algorithmically embeds it directly into the pixel data of the newly generated image. This goes beyond simple input-output; it’s about fusing one modality (text) invisibly inside another (image).
The user experience is about power. It enhances it by giving the user a sense of control and secrecy that a simple image-and-text app could never provide. The magic isn’t in seeing the two modalities work together; it’s in knowing that one is invisibly controlling the other. It’s a demonstration of how multimodal AI can be used for more than just cute chatbots and summary tools. It can be used to keep secrets.
This content originally appeared on DEV Community and was authored by Aman Choudhary