Text-to-Image-Phi2 Browser Extension

Created by team Metaverse Professional on February 02, 2024

This project introduces a groundbreaking browser extension designed to transform the way we interact with text on the internet. At its core, the extension empowers users to effortlessly select any piece of text they encounter online. With a simple right-click, the extension springs into action, harnessing the power of a context menu to capture the selected text. This text is then seamlessly sent to the cutting-edge Microsoft Phi-2 API, where it undergoes a concise summarization process. The resulting prompt, a distilled essence of the original text, is ingeniously crafted to capture its core message and themes. The magic doesn't stop there. This summarized prompt becomes the input for the state-of-the-art generative AI model, Dalle 3. Dalle 3, known for its exceptional capability to generate vivid, contextually rich illustrations, takes the prompt and brings it to visual life. The outcome is an illustration that not only embodies the essence of the selected text but also adds a layer of visual interpretation that enriches the user's interaction with the content. This extension is not just a tool but a bridge between the textual and visual realms, opening up new possibilities for understanding and engaging with digital content. Through its innovative use of AI technologies, it offers a unique, interactive experience that enhances the way we consume information on the internet.

Category tags: