In our search for a third-party animation solution, we found that no available options met our stringent criteria for fidelity, latency, speed, real-time emotion conveyance, and quality lip-sync. This led us to develop our own groundbreaking solution: an AI/Mocap (Motion Capture) hybrid, multi-layered animation system.

Our system uniquely differentiates between various parts of the face, with each segment animated using distinct techniques and then seamlessly layered together. This innovative approach combines the strengths of pre-made mocap facial animations and AI-driven emotional responsiveness. The result is extraordinarily lifelike facial expressions that dynamically match the mood of the conversation and the specific intentions of each phrase. For instance, when a question is asked, the avatars' expressions can shift to an inquisitive look, with variations depending on the conversational context.

Furthermore, the lip movements are procedurally generated in real-time to align perfectly with the AI-generated voice, enhancing the realism and engagement of Pete and Chloe.