← back
Give Your Chat Agent a Voice — Luke Harries, Head of Growth, ElevenLabs
Takeaway
Voice Engine lets you slap voice onto any existing chat agent in one prompt instead of replacing the whole stack.
Summary
- ElevenLabs is launching Voice Engine — a primitive that wraps any existing chat agent with their voice stack instead of forcing a full agent rebuild.
- Combines Scribe speech-to-text, V3 text-to-speech, emotion-aware turn-taking, thousands of voices, and shadcn/Vercel-style UI components.
- Server SDK is a thin wrapper that proxies to your existing chat agent; client SDK adds a widget in ~3 lines and gets telephony/CSAT for free.
- Tool calling stays in the existing chat agent — ElevenLabs adds client-side tools that can manipulate the DOM via the voice widget.
- Predicts chat agents will either die or add voice; voice unlocks omni-channel paradigms (e.g., a PostHog voice agent joining Zoom calls).
voiceelevenlabschat-agents
Original description
Chat agents dominated 2025. Every product either went AI-first or got left behind. But text-in, text-out is already starting to feel dated. Voice is faster, more accessible, and opens up interaction paradigms that chat just can't touch — phone lines, Zoom calls, screen readers, ambient interfaces. In this talk, Luke Harries from ElevenLabs argues that the next upgrade for every chat agent isn't better prompts or smarter RAG. It's a voice layer. The problem is most teams have already built and tuned their chat agents. They don't want to throw that out. This session shows how ElevenLabs' Voice Engine wraps any existing agent in a few lines — handling turn-taking, speech-to-text, text-to-speech, and emotion-aware interruption detection — without touching the underlying logic. There's a live demo of converting a working chat support agent to voice in a single prompt, plus a look at the client and server SDKs, Shadcn-based UI components, and how tool calling still works through the wrapper. Speaker info: - https://www.linkedin.com/in/luke-harries - https://harries.co/ Timestamps 0:00 Introduction to voice-first chat agents 0:20 The shift from text-based to voice-based interactions 1:43 Evolution of agent architecture and challenges of rebuilding 2:47 Introducing the ElevenLabs Voice Engine 3:32 Overview of the server and client SDKs 4:36 UI components and deployment demo 5:56 Summary of voice engine integration paradigms 6:37 Predictions for the future of AI agents 7:00 Q&A: Handling tool calling and integrations