🎙️ media2text (AI Audio & Video Transcriber)

Convert audio and video files into accurate text transcripts using AI (Whisper, DeepSeek, or OpenRouter models).
Supports multiple formats and can be self-hosted with minimal setup.

🚀 Features

• 🔊 Supports audio files: .mp3, .wav, .m4a, .ogg
• 🎥 Supports video files: .mp4, .mkv, .mov (extracts audio automatically)
• 🧠 Powered by AI models:
  • OpenAI Whisper
  • DeepSeek (via OpenRouter)
• 🔧 Self-hosted: run locally on your own machine
• 🗂 Outputs clean, timestamped text files

🛠 Installation

git clone https://github.com/kh4sh3i/media2text.git
cd media2text
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

Create your .env file (based on .env.example) and add your API keys if needed.

🧪 Usage

python media2text.py "path/to/media"

🧠 Supported Models

Model	API Support	Offline	Notes
Whisper	✅ Yes	✅ Yes	Best balance of accuracy/speed
DeepSeek	✅ Yes	❌ No	Requires API key via OpenRouter

Tip

for export audio from video file you sohuld install FFmpeg

📄 License

MIT License. Use freely, credit appreciated.