Meeting Summary Generator

This project is designed to help you easily create summaries of your group meetings using Python. It leverages the Whisper model for speech recognition and the Gemma model from Ollama for generating concise meeting summaries.

Table of Contents

• Features
• Requirements
• Installation
• Usage
• Examples
• License

Features

• Speech to Text: Converts audio recordings of meetings into text using the Whisper model.
• Text Summarization: Generates concise summaries of meeting transcriptions using the Gemma model.
• Flexible Output: Save the transcriptions and summaries to specified directories for easy access.

Requirements

• Python 3.12.4 or higher.
• Ollama serve models.
• Required Python packages:
  • speechrecognition
  • ollama
  • openai-whisper

Installation

1. Clone the repository:

   git clone https://github.com/franciscopalmeromoya/meeting-summaries.git
   cd meeting-summaries

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the required packages:

   pip install -r requirements.txt

4. Install Whisper and Gemma models:

   • Follow the official instructions for installing and setting up Whisper .
   • Install the Ollama Gemma model as per the instructions provided on the Ollama website.

Usage

Speech to Text

Convert an audio file to text using the speech2text.py script.

python speech2text.py -a path/to/your/audio/file.wav -o path/to/output/directory -m base

• -a: Path to the audio file (required).
• -o: Path to the output directory (optional).
• -m: Whisper model to use (default is base).

Generate Meeting Summary

Generate a summary from a text file using the gemma.py script.

python gemma.py -a path/to/your/transcription.txt -o path/to/output/directory -m general

• -a: Path to the text file (required).
• -o: Path to the output directory (optional).
• -m: Gemma prompt mode (default is general). Options are general and key-takeaways.

Examples

Example 1: Convert Audio to Text

python speech2text.py -a meeting_audio.wav -o transcriptions -m base

Example 2: Generate Summary from Transcription

python gemma.py -a transcriptions/meeting_audio.txt -o summaries -m key-takeaways

License

This project is licensed under the MIT License. See the LICENSE website for details.