Welcome to the VisionVerbalizer repository! This project showcases a Node.js application that leverages the Google Gemini API to generate captions for images, enhancing accessibility for visually-impaired users.
A. User Interaction: When the user selects an image and clicks the "Submit" button, the frontend code triggers the "getCaptionForImage" function.
B. Sending the Request: The "getCaptionForImage" function creates a FormData object, appends the image file to it, and makes a POST request to the backend endpoint /generate-caption using axios.
C. HTTP POST Request: The "axios.post" function sends an HTTP POST request. This is the endpoint on the backend where the image file will be sent. The Content-Type header is set to multipart/form-data, which indicates that the request contains file data.
"displayCaption": The displayCaption function updates the user interface with the returned caption or error message.
A. Powered by Node.js and Express, this application handles image uploads, communicates with the Google Gemini API to generate captions, and serves responses to the frontend.
B. Integration: Seamless connection between the frontend and backend with Axios for API requests, and Multer for file handling. Real-time caption generation and voice playback are supported.
Visit https://ai.google.dev/ to generate an API key for the Google Gemini API.
In the backend directory, create a file named .env and store your API key as follows: API_KEY=your-google-gemini-api-key
Open your browser and start using VisionVerbalizer. Upload any image you want to create captions for and Wohoo there is your Captions. You can Listen or Stop the caption sound anytime you want!!!
