Meeting Summary Generator

The Meeting Summary Generator is a Python-based tool that extracts audio from an MP4 video of a virtual meeting, transcribes the audio using OpenAI's Whisper model, and generates a summary of the meeting using OpenAI's GPT models. You can also start the process from an audio file or a text file.

Features

Extract audio from video files (MP4 format)
Transcribe audio using OpenAI's Whisper model
Summarize transcriptions using OpenAI's GPT-4o model
Save generated meeting summaries as text files

Requirements

Download Python 3.10 or higher (required for compatibility with certain libraries and features)
Get an OpenAI API key (keep this key secure and do not commit it to version control)

Installation

Clone this repository:

git clone https://github.com/samimhidia1/meeting-summary-generator.git

Change to the project directory:
```
cd meeting-summary-generator
```
Create a virtual environment:
```
python3 -m venv venv
```
Activate the virtual environment:
- if you are using Windows:
```
venv\Scripts\activate
```
- if you are using Linux or macOS:
```
source venv/bin/activate
```
Install the project dependencies:
```
pip install -r requirements.txt
```

Alternatively, you can use the provided setup script to automate the creation of the virtual environment and installation of dependencies:

./setup.sh

Usage

Quick Start Guide

Set up the OpenAI API key: Create a file named openai_apikey.txt in the root directory of the project and add your OpenAI API key to this file. The main.py script will read the API key from this file.
```
echo "your-openai-api-key" > openai_apikey.txt
```
Run the script for the first time: Run the main.py script in your terminal:
```
python main.py
```
The script will prompt you to enter the name of the project you want to generate a summary for. Enter the name of the project and press Enter.

e.g. Project MK-Ultra

If the project does not exist, the script will create a new project directory with the name you entered. The project directory will be created in the projects directory.

Detailed Usage Instructions

Adding a video file to an existing project

Place the video file you want to process in the videos directory of the project directory.
The script will extract the audio from the video file and save it in the audios directory of the project directory.
The script will transcribe the audio and save the transcription in the transcriptions directory of the project directory.
Finally, the script will generate a summary of the transcription and save it in the summaries directory of the project directory.

Adding an audio file to an existing project

Place the audio file you want to process in the audios directory of the project directory.
The script will transcribe the audio and save the transcription in the transcriptions directory of the project directory.
Finally, the script will generate a summary of the transcription and a structured meeting summary from it and save both text files in the summaries directory of the project directory.

Adding a text file to an existing project

Place the text file you want to process in the transcriptions directory of the project directory.
The script will generate a summary of the transcription and a structured meeting summary from it and save both text files in the summaries directory of the project directory.

Running the script for an existing project after adding a video file, audio file, or text file

Once you have placed the video file, audio file, or text file in the appropriate directory, re-run the main.py script in your terminal:
```
python main.py
```
Select the name of the project you want to generate a summary for and press Enter.
The script will then prompt you to choose between the following options:
- 1: Start from video file
- 2: Start from an audio file
- 3: Start from a text file
- 4: Exit
Choose the option that best suits your needs and press `Enter.

Option 1: Start from video file

If you choose option 1, the script will prompt you to enter the name of the video file you want to extract the audio from. Enter the name of the video file with the extension and press Enter.

e.g. meeting.mp4

Option 2: Start from an audio file

If you choose option 2, the script will prompt you to enter the name of the audio file you want to transcribe. Enter the name of the audio file with the extension and press Enter.

e.g. meeting.wav

Option 3: Start from a text file

If you choose option 3, the script will prompt you to enter the name of the text file you want to summarize. Enter the name of the text file with the extension and press Enter.

e.g. meeting.txt

Option 4: Exit

If you choose option 4, the script will exit.

Structure of the generated summary

The generated summary is structured as follows:

STRUCTURE:
1. Meeting Objectives:
    a. Objective 1: [Brief description]
    b. Objective 2:
    .
    .
    .
    c. Objective n:
2. Key Discussion Points:
    a. Topic 1: [Brief summary of discussion]
        Main point 1
        Main point 2
        ...
        Main point n
    b. Topic 2:
        Main point 1
        Main point 2
        ...
        Main point n
    .
    .
    .
    n. Topic n:
        Main point 1
        Main point 2
        ...
        Main point n
3. Important Decisions:
    a. Decision 1: [Decision made and rationale]
    b. Decision 2:
    .
    .
    .
    n. Decision n:
4. Action Items:
    a. Action Item 1: [Task description] - [Deadline]
    b. Action Item 2:
    .
    .
    .
    n. Action Item n:
5. Meeting Outcomes:
    a. Objective 1: [Status - Achieved/Partially Achieved/Not Achieved]
        [Key takeaways, insights, or progress made]
    b. Objective 2:
    .
    .
    .
    n. Objective n:
6. Next Steps:
    a. Next Meeting Date: [Next Meeting Date]
    b. Agenda Items for Next Meeting: [List of Agenda Items]
    c. Additional Follow-up: [Any other relevant information or follow-up tasks]

You can change the structure of the summary by editing the summary_structure.txt file in the generate_meeting_summary/prompts directory and the line 111 of the pipelines.py script.

Project Structure

The project is structured as follows:

meeting_summary_generator/
├── audio_extractor
│   └── audio_extractor.py
├── generate_meeting_summary
│   ├── generate_meeting_summary.py
│   └── prompts
│       ├── summary_structure_2.txt
│       └── summary_structure.txt
├── main.py
├── meeting_summarizer
│   ├── meeting_summarizer.py
│   ├── prompts
│   │   └── summarize_transcript.txt
│   └── utils.py
├── openai_api_interaction
│   └── openai_api_interaction.py
├── pipelines.py
├── projects
│   └── MK-Monarch
│       ├── audios
│       ├── summaries
│       ├── transcriptions
│       └── videos
├── README.md
├── requirements.txt
├── speech_transcriber
│   └── speech_transcriber.py
├── temp
└── tests
    ├── test_audio_extractor.py
    ├── test_meeting_summarizer.py
    └── test_speech_transcriber.py

Testing

You can run the unit tests with the following command:

python -m unittest discover  tests

Note: Running the tests for speech_transcriber and meeting_summarizer will consume tokens from your OpenAI API quota, so use it judiciously to avoid running out of your allocated tokens.

Setup Script

The setup.sh script automates the creation of the virtual environment and installation of dependencies. To use the script, run the following command:

./setup.sh

The script will create a virtual environment, activate it, and install the required dependencies from the requirements.txt file.

TODO

Add support for other video formats
Add support for other audio formats
Rewrite the tests to use mocks instead of the actual OpenAI API
Add support for other Transformers models
Add support for other prompts and summary structures
Add support for other languages and automatic language detection
Make the text in the terminal more user-friendly
Test for other operating systems

Acknowledgements

License

MIT

Contact

Sami Mhidia - sami.mhidia@jarvisator.com