WhisperForge

WhisperForge is an AI-powered audio transcription and analysis tool that integrates OpenAI's Whisper model with Notion for documentation.

Features

• Audio file transcription (MP3, WAV, M4A, OGG)
• Automatic chunking of large audio files
• Language detection and support
• Direct export to Notion
• Insights extraction from transcribed content

Prerequisites

• Python 3.11 (required for audioop support)
• ffmpeg
• OpenAI API key
• Notion API key and integration

Installation

1. Create Python virtual environment:

python3.11 -m venv whisperforge-

2. Install dependencies:

pip install -r requirements.txt

3. Install ffmpeg:

# macOS
brew install ffmpeg

# Ubuntu/Debian
sudo apt-get install ffmpeg

Configuration

1. Create .env file with your API keys:

OPENAI_API_KEY=your_openai_key_here
NOTION_API_KEY=your_notion_key_here

2. Set up Notion integration:
   • Create a new integration at https://www.notion.so/my-integrations
   • Share your target Notion page with the integration

Usage

Run the Streamlit app:

streamlit run app.py

Contributing

[Your contribution guidelines here]

License

[Your chosen license here]

Development Status

Currently implementing:

• Notion integration for permanent storage
• Monitoring and observability with Prometheus/Grafana
• Service health checks and logging