Receipt and Refund Request Tracker

A comprehensive application for managing receipts and processing refund requests, featuring AI-powered audio and image processing capabilities.

Features

Receipt Processing

• Upload and store receipt images
• Extract amount information from receipts
• Store receipt data in Supabase database

Audio Processing

• Process audio files from refund requests using Groq's whisper-large-v3-turbo model
• Automatic transcription of audio content
• AI-powered summarization of transcriptions
• Support for direct URL audio file processing

Database Integration

• Supabase backend for secure data storage
• Real-time data synchronization
• Efficient query operations

Setup

1. Install required packages:

pip install streamlit supabase langchain-groq requests python-dotenv

2. Set up environment variables in .env:

GROQ_API_KEY=your_groq_api_key
SUPABASE_URL=your_supabase_url
SUPABASE_KEY=your_supabase_key

3. Launch the application:

streamlit run app.py

Usage

Processing Refund Requests

1. Upload receipt images through the Streamlit interface
2. Record or upload audio explanations for refund requests
3. View processed results including:
   • Extracted receipt amounts
   • Audio transcriptions
   • AI-generated summaries

Audio Processing Features

• Automatic download and processing of audio files
• High-accuracy transcription using Groq's whisper model
• Concise summaries of audio content
• Batch processing capabilities
• Comprehensive error handling

Technical Details

Database Schema

• refund_requests table:
  • id: Primary key
  • amount: Numeric (receipt amount)
  • image_url: Text (receipt image location)
  • audio_url: Text (refund request audio location)

Key Components

1. app.py: Main Streamlit application
2. audio_processor.py: Audio processing functionality
3. database_operations.py: Supabase database interactions
4. receipt_processor.py: Receipt image processing

Error Handling

• Robust error management for file operations
• Graceful handling of processing failures
• Detailed error reporting
• Automatic cleanup of temporary files

Security

• Secure handling of API keys through environment variables
• Safe temporary file management
• Secure URL processing
• Protected database operations

Requirements

• Python 3.7+
• Streamlit
• Supabase
• langchain-groq
• requests
• python-dotenv

Support

For issues and feature requests, please create an issue in the repository.

Hackfest