Low-Latency 3D AI Characters in Unity with Real-Time Lip Sync | Convai Unity Tutorial

Hello everyone,

We are back with a comprehensive new tutorial for our Unity developers! If you are looking to build fully interactive AI characters with lifelike facial animation and incredibly low latency, this step-by-step guide is for you.

In this video, we walk you through how to set up an interactive AI character in Unity using Convai’s updated WebRTC-based Unity plugin.

:sparkles: What Makes This Update Special?

Featuring our demo character Susan, this tutorial showcases how you can achieve natural, fluid conversations inside Unity. The updated plugin enables:

  • :high_voltage: WebRTC Voice Interaction: Experience significantly lower response latency for fluid turn-taking and realistic exchanges.

  • :speaking_head: Hands-Free Conversations: Utilize Voice Activity Detection (VAD) for natural dialogue without push-to-talk inputs.

  • :mouth: NeuroSync Integration: High-quality, real-time facial animation and accurate lip sync powered by Convai’s in-house AI model.

  • :brain: Deep Context: Build smarter characters with long-term memory and multimodal knowledge integration.

:hammer_and_wrench: What You Will Learn

We cover the entire pipeline from scratch so you can easily integrate lifelike AI avatars into your games, training simulations, and XR experiences:

  1. Installation: How to install the Convai SDK via the Unity Package Manager (UPM) or the Unity Asset Store.

  2. Core Setup: Configuring API keys, the Convai Manager, Player, and Character components.

  3. UI Integration: Enabling a transcript UI for in-scene conversations.

  4. Facial Animation: Setting up real-time lip sync with NeuroSync on a custom avatar.

:link: Resources & Downloads

Ready to build your own intelligent, lip-syncing avatars? Grab the plugin and check out the documentation below:

If you run into any issues during setup or want to share your completed projects, let us know in the replies below.

Happy developing!

1 Like