ChatdollKit

What is ChatdollKit?

ChatdollKit is an open-source Unity framework engineered for creating interactive 3D virtual agents, VTubers, and autonomous AI companions. It bridges the gap between modern generative AI models and real-time 3D animation systems, delivering seamless, natural interactions across desktop, mobile, and XR platforms.

End-to-End Multimodal Coordination

At its core, ChatdollKit orchestrates the entire conversational pipeline: speech recognition (STT), natural language processing via Large Language Models (LLMs), text-to-speech generation (TTS), and expressive physical performance. It automatically synchronizes voice-driven lip-syncing, realistic blinking, dynamic facial expressions, and contextual body gestures according to the dialogue context.

Key Applications of ChatdollKit:

• Autonomous AI VTubers: Power 24/7 interactive livestreams capable of chatting with fans and reacting to real-time events.
• Immersive Virtual Companions: Build engaging cross-platform companions for mobile, PC, and VR headsets.
• Digital Kiosks & Smart Receptionists: Deploy visually compelling interactive brand ambassadors and virtual concierges.

Thanks to its highly modular architecture, developers can swap between cloud and local AI services effortlessly and expand conversational flows directly within Unity.