Multilanguage Invoice Extractor | Gemini Flash Model
This project is a Streamlit-based web app that extracts information from invoices using Google's Gemini LLM model. Users upload an invoice image, and the app processes it alongside a text prompt to generate a response based on the content of the invoice. The app leverages the Google Gemini API for content generation and displays the AI's response on the interface. It supports image uploads (JPG, PNG, JPEG) and provides detailed insights by analyzing the uploaded invoice.
This application leverages Google's Gemini 1.5 Flash model to extract and interpret data from invoice images. It supports multiple languages and integrates environmental variables and a Streamlit interface to interact with users, providing real-time invoice analysis.
Features
- Environment Variables: Manages sensitive data securely with Python's dotenv package.
- Streamlit Interface: Provides a user-friendly, web-based interface for uploading and analyzing invoices.
- Image Processing: Uses the PIL library to handle image uploads and processing.
- Multilanguage Support: Can extract and interpret invoice information in multiple languages.
- Gemini 1.5 Flash Model: Utilizes the advanced AI capabilities of the Gemini 1.5 Flash model for accurate content generation based on the image and text inputs.
Prerequisites
- Python 3.8+
- Streamlit
- dotenv Python package
- PIL (Pillow for image processing)
- os
- genai
Usage
- Upload an invoice image using the file uploader.
- Use the text input box to input any specific queries related to the invoice.
- Click the "Tell me about the invoice" button to process the image and receive detailed information.
