Duck E

DUCK-E: The Duck That Talks Back

**DUCK-E** (Digitally Unified Conversational Knowledge Engine) is an AI-powered voice assistant inspired by rubber duck debugging. Instead of explaining your problem to a silent rubber duck, DUCK-E listens and talks back — asking questions, offering suggestions, and searching the web in real time.

How it works

Browser (WebRTC) → WebSocket → FastAPI → OpenAI Realtime API
                                              (gpt-realtime-2)

1. Your browser captures audio via the MediaDevices API and opens a WebRTC peer connection.
2. The FastAPI backend requests an ephemeral key from OpenAI — your real API key never touches the browser.
3. Speech is transcribed by OpenAI Whisper-1; the Realtime API handles both understanding and voice response natively.
4. When DUCK-E needs external data (weather, web search), it calls the appropriate tool on the server and folds the result into its reply.

Features

Voice I/O

• Low-latency, full-duplex voice conversation via the OpenAI Realtime API
• 11 built-in voices — changeable mid-session without reconnecting
• Interruption-friendly: DUCK-E handles natural conversation flow

Tools

Tool	What it does
get_current_weather	Current conditions via Open-Meteo (free, no key required)
get_weather_forecast	Multi-day forecast via Open-Meteo
web_search	Live web search via OpenAI's gpt-4o-mini + web_search_preview
web_fetch	Fetches and parses a URL (SSRF-protected; blocks private IPs)
save_memory / recall_memories	Stores and retrieves facts about the user
change_voice	Switches voice mid-session

Persistent memory

DUCK-E stores per-user facts with categories, confidence scores, and time decay, and surfaces them at the start of each session. Memory is keyed by user identity via:

• Google OAuth (recommended): Users sign in with Google, memory is keyed to their email
• Reverse proxy headers: Falls back