This is a Python script that uses a local LLM model via Ollama to summarize meeting minutes. It extracts key discussion points, decisions made, action items, and follow-up tasks.

• Uses a local LLM model (e.g., mistral, gemma, llama3, deepseek).
• Summarizes meeting minutes into key points and action items.
• Outputs the summary on screen and saves it to a file.
• Lightweight, requires only ollama as a dependency.

Ollama is required to run the local LLM model. Install it from the official website:

🔗 Download Ollama

After installation, verify it is working by running:

ollama --version

ollama pull mistral   # Small and fast model (default)
ollama pull gemma     # Another lightweight model
ollama pull llama3    # More powerful model
ollama pull deepseek  # Largest and most powerful model

pip install ollama

python3 ./meeting_summarizer.py meeting_minutes.vtt

By default, the script uses the mistral model. To use a different model, specify it as an argument:

python3 ./meeting_summarizer.py meeting_minutes.vtt --model gemma

The summary will be printed on screen. A summary file will be saved as meeting_minutes_summary.txt.