Video Captioning System with AI Transcription

Installation

Prerequisites

• Python 3.8 or higher
• CUDA-capable GPU (recommended for faster processing)
• Ubuntu/Debian system or WSL for Windows users

Automatic Installation

Clone the repository:

git clone https://github.com/Gitanshu1903/Educational-Video-Processor.git
cd Educational-Video-Processor

Manual Installation

Install Python dependencies:

# Install Python dependencies:
pip install faster-whisper -q
pip install moviepy==2.0.0.dev2 -q
pip install imageio==2.25.1 -q

# Install system dependencies (Ubuntu/Debian):
sudo apt-get install libcublas11 -y
sudo apt install imagemagick -y

# Configure ImageMagick permissions:
!cat /etc/ImageMagick-6/policy.xml | sed 's/none/read,write/g'> /etc/ImageMagick-6/policy.xml

Quick Start

1. Place your video file in the project directory.
2. Run the script:

python src/main.py

3. Find your captioned video as output.mp4 and content summary as summary.json.

Configuration

Caption Styling

style = CaptionStyle(
    font_name="Helvetica",
    font_size=32,  # Auto-adjusts based on video height
    text_color="white",
    highlight_color="yellow",
    stroke_color="black",
    stroke_width=1.5,
    background_color=(64, 64, 64),
    background_opacity=0.6
)

Video Processing Settings

spec = VideoSpec(
    fps=24,
    codec="libx264",
    audio_codec="aac",
    threads=4,
    bitrate="8000k",
    preset="medium"
)