This project features an Invoice Extraction Bot that leverages Gemini, a large language model developed by Google, to extract information from multiple images of invoices. The bot is implemented in Python and integrates with Streamlit for the frontend interface.
-
Invoice Information Extraction: The bot utilizes Gemini to analyze and extract relevant information from images of invoices.
-
Multiple Image Support: Users can upload multiple images of invoices, and the bot processes each image to extract necessary details.
-
Question-Answering System: The bot provides answers to questions related to the invoices, such as total amount, due date, vendor details, etc.
-
Streamlit Frontend: The frontend interface is built using Streamlit, offering a user-friendly experience for uploading images and viewing extracted information.
To run the Invoice Extraction Bot locally, follow these steps:
-
Clone the repository:
git clone https://github.com/ShivankK26/Invoice-Extraction-Bot.git
-
Install the required dependencies:
pip install -r requirements.txt
-
Start the Streamlit server by running the following command:
streamlit run app.py
-
Open your web browser and navigate to the provided URL to access the Streamlit interface.
-
Upload one or multiple images of invoices using the provided interface.
-
Once the images are uploaded, the bot will process them using Gemini and display the extracted information.
-
Users can ask questions related to the invoices in the provided text box, and the bot will provide answers based on the extracted data.