Speech-to-Speech

Voice conversation without visual rendering.

What is it?

A real-time voice conversation with an AI character — listen, think, speak — without on-screen avatar rendering. Useful for voice-only products and telephony-style flows.

When to use it

  • Voice assistants and phone integrations
  • Products where audio is the entire interface
  • Low-bandwidth or audio-only environments
  • When you do not need lip-sync or facial animation

API

import { Experience } from '@liforma/client';

const experience = await Experience.startSession({
  experienceId: 'exp_…',
  render: false
});

experience.on('message', ({ role, text }) => {
  console.log(role, text);
});

await experience.attach({ container: null });

When to use Avatar Experiences instead

Use a full Avatar Experience when the character should appear on screen with real-time animation. Speech-to-Speech is the voice layer without visuals; Avatar Experiences combine voice, intelligence, and animation in one integration.