How I used AI voice to help me play Magic: The Gathering
summary
Built a voice assistant that answers Magic the Gathering rules and guides play by piping speech-to-text → LLM → text-to-speech, using Twilio Conversation Relay to connect a phone number to a Python app, an OpenAI model labeled GBT 40 mini for reasoning, and 11Labs voices (with Deepgram available for STT). The system stores session context, asks follow-ups, and explains mechanics; a live demo’s microphone failed but a prerecorded demo verifies the assistant works and the code is public.
AI-written summary · about
chapters (7)
- 0:00AI voice and Magic
- 10:01Human vs computer dev
- 24:52Voice AI challenges
- 37:21Jace AI assistant
- 44:27Live demo troubleshooting
- 55:11Conversation Relay powering AI
- 1:00:12Future of AI voice
related talks
51mBuilding an AI assistant that makes phone calls
17mBuilding Effective Voice Agents
01h 02m[Full Workshop] Building Conversational AI Agents
44mSmartly AI: Cross Platform Voice Application Creation
10mMeet the Speakers
01h 26mBuilding voice agents with OpenAI
21mMeet your new and improved teammate! Human-directed AI agents and gaming
01h 04mBuilding Expertise Into Generated Conversations
