Vishy is a web app that generates a description from an image, allowing those who are visually impaired to know what an image is describing, even if no alternate text exists for the image.
To run Vishy, you need to have a Gemini API Key. Once you downloaded this repository, add a .env file Then, open a terminal, and run the command uvicorn main:app --reload. Then, open index.html, and upload an image, and then describe an image.