Video Caption Generator

A Python script that automatically adds animated captions to videos using speech recognition, with extensive customization options.

Features

• Automatic Captioning

  • Uses OpenAI's Whisper model for speech-to-text transcription
  • Word-by-word or multi-word animated captions
  • Customizable zoom animation effects

• Styling Options

  • Color cycling for captions
  • Configurable font, size, and positioning
  • Text stroke and background options
  • Highlight individual words

• Advanced Video Editing

  • Optional intro sound
  • Video overlay support
  • Flexible caption placement

Prerequisites

• Python 3.8 or higher
• FFmpeg installed on your system
• GPU recommended for faster processing (optional)

Installation

1. Clone the repository:

git clone <repository-url>
cd video-caption-generator

2. Install dependencies:

pip install moviepy whisper PyYAML requests fonttools pillow

Font Management with fontify.py

The fontify.py script provides an easy way to download fonts from Google Fonts:

# Download a specific font
python fontify.py "Open Sans"
# Or
python fontify.py "Roboto Black"

This script will:

• Query the Google Fonts API
• Download the font file
• Save it in a ./fonts directory

Configuration

Configuration is managed through a config.toml file with numerous options:

Caption Display

• number_of_words: Number of words to display at once
• font: Font filename
• font_size: Text size in pixels
• position: Screen position (top/center/bottom)
• text_align: Text alignment
• text_colors: Color cycling for captions
• stroke_color and stroke_width: Text outline styling
• bg_color: Background color or transparency

Advanced Options

• transition: Zoom animation toggle
• highlight: Highlight current word
• has_intro_sound: Add background audio
• overlay: Overlay another video

Usage

Command Line

# Basic usage
python caption.py input_video.mp4

# Specify output and configuration
python caption.py input_video.mp4 --output_file captioned_video.mp4 --config custom_config.toml

Programmatic Usage

from caption import add_captions_to_video

add_captions_to_video(
    "input_video.mp4", 
    output_path="output_video.mp4", 
    config_path="config.toml"
)

Video Silence Trimmer

Included silence_trimmer.py allows removing silent segments from videos:

python silence_trimmer.py input_video.mp4

Troubleshooting

1. Font Issues

   • Use fontify.py to download missing fonts
   • Ensure fonts are in the ./fonts directory

2. Performance

   • Use smaller Whisper models for faster processing
   • Reduce video resolution if memory is limited

3. Dependencies

   • Ensure all required libraries are installed
   • Check FFmpeg installation

Customization Tips

• Experiment with config.toml settings
• Try different fonts using fontify.py
• Adjust Whisper model for accuracy vs. speed

Technologies Used

• MoviePy: Video processing
• Whisper: Speech recognition
• PIL (Pillow): Image manipulation
• TOML: Configuration management

License

• Apache License 2.0 (for Roboto Fonts)
• MIT License for the script

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a pull request

Credits

• OpenAI (Whisper)
• Google Fonts
• MoviePy Developers