Skip to content

Repo containing the code to build an AI meeting assistant that schedules meetings based on the user's request

Notifications You must be signed in to change notification settings

gabriel1628/Meeting-Scheduling-Assistant

Repository files navigation

LLM-Based Meeting Scheduling Assistant

Overview

The LLM-Based Meeting Scheduling Assistant is a sophisticated tool designed to streamline the process of scheduling meetings. Powered by a Large Language Model (LLM), this assistant automates scheduling tasks by understanding natural language requests, analyzing your calendar, and suggesting suitable meeting slots.

Features

  1. Natural Language Understanding:
    The assistant takes natural language requests as input, such as:
    "Schedule a meeting with Mohamed next week."

  2. Calendar Integration:
    It integrates seamlessly with your calendar to check your availability and identify possible meeting slots.

  3. Smart Slot Suggestions:
    Based on the request constraints, the assistant suggests three possible time slots that work for you.

  4. Interactive Confirmation:
    You can select your preferred time slot from the suggestions.

  5. Automated Event Creation:
    The assistant creates a calendar event with a clear and descriptive title, e.g., "Meeting with Mohamed", ensuring your calendar stays organized.

Getting Started

Prerequisites

  • Python 3.10
  • Access to the Google Calendar API

Installation

  1. Clone the Repository:

    git clone https://github.com/yourusername/llm-meeting-scheduler.git
    cd llm-meeting-scheduler
  2. Install Dependencies:

    pip install -r requirements.txt
  3. Set Up the Google Calendar API:

    • Follow the instructions in the docs/GoogleCalendarAPI_Setup.md file.
    • Place your credentials file (e.g., calendar_credentials.json) in a safe place.
    • Modify the file_path variable that indicates the path to your credentials in the relevant notebooks.

Usage

  1. Run the Assistant (not yet available):

    python main.py
  2. Input Meeting Request:

    • Enter a meeting scheduling request, such as "schedule a meeting with Mohamed next week."
  3. Choose a Time Slot:

    • The assistant will suggest three time slots based on your calendar. Choose one of them.
  4. Event Creation:

    • The assistant will create a calendar event with a suitable title, such as "Meeting with Mohamed."

Configuration

  • Time Zone:
    Set your time zone in the config.json file to ensure proper slot suggestions.

  • Meeting time slots: By default, the assistant suggests meeting slots between 8:00 AM and 6:00 PM. You can customize this range in the config.json file.

  • Default Event Duration:
    The assistant defaults to a 30-minute meeting duration. You can adjust this in the config.json file.

  • LLM Model Configuration:

    • Update the config.json file with your LLM model configuration.

About

Repo containing the code to build an AI meeting assistant that schedules meetings based on the user's request

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published