Meeting Analyzer with AI

This project provides a Streamlit application for analyzing and transcribing meeting audio using AI. You can either upload a pre-recorded meeting or record a new one directly from the application. The audio is then processed, transcribed, and analyzed to generate a meeting summary.

Features

1. Summary of the Meeting: Provides a concise overview of the main points and action items discussed during the meeting.
2. Minutes of the Meeting: Generates a detailed transcript of the meeting, including time-stamped notes of what was discussed.
3. Analysis Visualization: Visualizes key metrics and insights from the meeting.

Architecture

The application is split into two main components:

1. Frontend: A Streamlit application that handles user interactions and displays results.
2. Backend: A FastAPI server that processes audio files and performs the analysis.

Usage

To run the application locally:

1. Ensure you have Docker and Docker Compose installed.
2. Clone this repository.
3. Navigate to the project directory and run:

   $env:DOCKER_BUILDKIT=1
   docker-compose up --build
   

4. Open a web browser and go to http://localhost:8501 to access the Streamlit interface.

Configuration

The application uses a configuration file to store settings like the POD URL. The configuration is loaded at startup and can be updated through the application.

Components

Frontend (Streamlit)

• GPU Utilization: Choose between "Local" and "API".
• LLM Inference Type: Select "Local LLM Model" or "Cloud LLM Model".
• Language: Choose between English and Arabic for the minutes of the meeting.
• Upload Meeting: Upload a pre-recorded meeting audio file.
• Record Meeting: Record a new meeting audio.
• Analyze Meeting: Analyze the uploaded or recorded meeting audio.

Backend (FastAPI)

• Audio Processing: Prepares and analyzes audio files.
• Configuration Management: Handles saving and loading of configuration settings.
• Output Generation: Creates formatted output for the frontend to display.

Code Structure

Frontend

• app.py: Main Streamlit application file.
• my_upload.py: Contains the custom upload component.
• my_recorder.py: Contains the custom record component.
• utils/style.py: Contains the function to apply styles to the Streamlit app.

Backend

• main.py: FastAPI server main file.
• utils/audio_utils.py: Contains functions to prepare audio files for analysis.
• utils/audio_analysis.py: Contains the function to analyze audio files.
• utils/output_utils.py: Contains functions to display the output of the analysis.
• utils/utils.py: Contains functions to save and load configuration settings.

Deployment

The application is containerized using Docker, making it easy to deploy in various environments. The docker-compose.yml file defines the services for both the frontend and backend.