media-to-text 🎥 ➡️ 📝

AI-Powered Media Transcription Using OpenAI's Whisper

---

🎯 Overview

A command-line tool that uses AI to automatically:

• Convert any audio/video to text with high accuracy
• Split large files into 25MB chunks (OpenAI's limit)
• Show real-time progress with elegant terminal output
• Handle errors gracefully with automatic cleanup
• Combine multiple transcripts into a single file

This is a basic, open-source setup designed to be easily:

• Forked and modified for your specific needs
• Integrated into larger projects
• Extended with additional features
• Shared and improved by the community

Feel free to use, modify, and share this tool as you see fit!

⚡️ Quick Start

1. Set Up Environment

# Clone the repository
git clone https://github.com/OkhDev/media-to-text.git
cd media-to-text

# Create virtual environment (REQUIRED)
python -m venv venv

# Activate virtual environment
# On macOS/Linux:
source venv/bin/activate
# On Windows:
# venv\Scripts\activate

# Install dependencies (after activating virtual environment)
pip install -r requirements.txt

2. Add Your API Key

# The script will create .env for you
# Just add your OpenAI API key:
OPENAI_API_KEY=your_key_here

3. Run It

# Make sure your virtual environment is activated
# You should see (venv) in your terminal prompt

# Transcribe media files
python transcribe.py

# Combine transcripts (optional)
python combine_transcripts.py

📁 Supported Formats

Video Formats 🎥	Audio Formats 🎵
.mp4 .mkv	.mp3 .wav
.webm .avi	.flac .aac
.mov .wmv	.m4a .ogg
.flv .m4v	.opus .wma
.3gp	.aiff .amr

📦 Requirements

• Python 3.8+
• FFmpeg
• OpenAI API key
• Virtual Environment
• Required packages:

  openai>=1.3.5
  python-dotenv>=1.0.0
  moviepy>=1.0.3
  httpx>=0.24.1
  

💡 Pro Tips

For Best Results:

• Use clear audio with minimal background noise
• Ensure sufficient disk space for temporary files
• Monitor your OpenAI API usage/costs
• Always use a virtual environment to avoid dependency conflicts

File Processing:

• Larger files are automatically split into chunks
• Each chunk must be under 25MB (OpenAI limit)
• Processing time depends on file size

Transcript Combination:

• All transcripts are saved in the transcripts/ directory
• Use combine_transcripts.py to merge multiple transcripts
• Combined file includes headers and separators for clarity
• Output includes timestamp for easy tracking

📂 Project Structure

media-to-text/
├── transcribe.py         # Main transcription script
├── combine_transcripts.py # Transcript combiner
├── requirements.txt      # Dependencies
├── .env                 # API key
├── media-files/         # Input files
├── transcripts/         # Individual transcripts
├── temp/               # Processing files
└── venv/               # Virtual environment

📝 Changelog

Version 1.0.1 (2024-03-17)

• Added virtual environment setup instructions
• Improved documentation readability
• Enhanced error handling for environment setup
• Updated dependency management approach

Version 1.0.0 (2024-03-16)

• Initial release
• Basic transcription functionality
• Support for multiple media formats
• Chunk processing for large files
• Transcript combination utility

🤝 Roadmap

Check out our Future Updates & Enhancements document for planned features and improvements.

🤝 Contributing

Found a bug or want to contribute? Feel free to:

• Open an issue
• Submit a pull request
• Suggest improvements

📄 License

MIT License - Use it, modify it, share it.

---

Made with ❤️ by OkhDev