Building Real-Time AI Voice Agents with LiveKit's Ben Cherry

The Neuron: AI Explained

Listen on: Spotify | Apple Podcasts | Podlink (1 h 9 m)

Topics: AI | Product | Technology

Ben Cherry from LiveKit shows how real-time voice agents work, from latency-sensitive pipelines to production-ready WebRTC infrastructure.

●	He explains why voice agents usually combine speech-to-text, an LLM, and text-to-speech.

●	He shows why WebRTC beats WebSockets for mobile, real-time calls under changing network conditions.

●	He demonstrates LiveKit’s builder, observability, and voice cloning for shipping and debugging agents.

"Voice is the interface for intelligence" - Ben Cherry