MAESTRO

MAESTRO (Multi-Agent Educational System for Training and Online Learning) is an AI-powered system designed to automate the creation of engaging educational video lectures. Leveraging Large Language Models (LLMs) and Text-to-Speech (TTS) technologies, MAESTRO offers two distinct pipelines for generating video content:

• Image-to-Video: Transforms existing slide decks (PDF presentations) into narrated video lectures, complete with instructor-style voiceover and synchronized visuals.
• Text-to-Video: Generates animated educational videos from text-based lesson outlines, utilizing AI agents to create dynamic slides and accompanying narrations.

Sample videos

Sample videos

Text-To-Video_KMeans.mp4

Image-To-Video_K-Means.mp4

Text-To-Video_KNN_TTS-1.mp4

You can find and download these and the rest of the sample videos generated by MAESTRO to see the system in action in the media directory.

media directory structure

The media directory comprises the following subdirectories with the corresponding files:

• Pipeline I: media/Image-To-Video (K-Means_TTS-1.mp4, K-Means_Clone.mp4, KNN_TTS-1.mp4, KNN_Clone.mp4)
• Pipeline II: media/Text-To-Video (K-Means_TTS-1.mp4, K-Means_Clone.mp4, KNN_TTS-1.mp4, KNN_Clone.mp4)

How to use source code

Prerequisites:

• Python 3.x

• Required Python Packages: (List them here - you can get this from your requirements.txt if you have one, or by listing key libraries like openai, fitz, pptx, ffmpeg-python, Flask, etc.)

  pip install -r requirements.txt

• OpenAI API Key: You will need an OpenAI API key to use the LLM and TTS functionalities.

To use Image-To-Video through terminal:

1. Navigate to the src/Image-to-Video directory:

   cd src/Image-to-Video

2. Run the video_generator.py script:

   python video_generator.py path/to/your/presentation.pdf your-openai-api-key --output output_video.mp4

   • Replace path/to/your/presentation.pdf with the actual path to your PDF slide deck.
   • Replace your-openai-api-key with your OpenAI API key.
   • --output output_video.mp4 specifies the desired output filename (optional, defaults to slideshow.mp4).

To chat with the content (after generating Image-to-Video):

1. Navigate to the src/Image-to-Video directory:

   cd src/Image-to-Video

2. Run the chat.py script:

   python chat.py output/scripts your-openai-api-key

   • output/scripts assumes you have already generated scripts using video_generator.py and they are located in the output/scripts directory (default behavior of video_generator.py). Adjust the path if necessary.
   • Replace your-openai-api-key with your OpenAI API key.

   This will launch a simple command-line interface for chatting with the content of your generated lecture.