Skip to content

QUICKSTART.md

Latest commit

 

History

History
75 lines (53 loc) · 1.68 KB

QUICKSTART.md

File metadata and controls

75 lines (53 loc) · 1.68 KB

Quick Start Guide

🚀 Get Started in 3 Steps

1. Setup (One-time)

# Install uv (if needed)
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install all dependencies
cd /home/wut/playground/reith-lecture
uv sync

# Configure your Google AI API key
cp .env.example .env
# Edit .env and add your GOOGLE_AI_API_KEY

2. Run the App

uv run python app.py

Open browser to: http://localhost:7860

3. Use the Interface

Tab 1 - Download:

  • Enter RSS feed: https://podcasts.files.bbci.co.uk/p00fzl9g.rss
  • Set limit: 3
  • Click "Download from RSS"

Tab 2 - Transcribe:

  • Refresh audio list
  • Select a downloaded file
  • Choose model: base (recommended)
  • Click "Transcribe Selected File"

Tab 3 - Chat:

  • Click "Load Transcripts to Vector Store"
  • Ask: "What are the main themes discussed?"
  • Get AI-powered responses with source citations!

📚 Popular BBC Feeds

  • Reith Lectures: https://podcasts.files.bbci.co.uk/p00fzl9g.rss
  • In Our Time: https://podcasts.files.bbci.co.uk/p01f0vzr.rss
  • The Documentary: https://podcasts.files.bbci.co.uk/p02nq0lx.rss
  • Analysis: https://podcasts.files.bbci.co.uk/b006r4vz.rss

💡 Tips

  • Start with base Whisper model (good balance of speed/accuracy)
  • Transcription is FREE and runs locally (no API costs)
  • Load transcripts to vector store once per session
  • All data stored locally in downloads/, transcripts/, and data/

🔧 Troubleshooting

Chat not working?

Need FFmpeg?

# Ubuntu/Debian
sudo apt install ffmpeg

# macOS
brew install ffmpeg