Can I use my own face and answer voice to generate a livestreaming?

Hello, I would like to know if it is possible to achieve the following functionality through the D-ID platform:

We aim to create a real-time interactive system for user engagement. When users ask questions, the system will respond based on a model we provide, using our pre-recorded audio files as the response voice. Additionally, it should use a custom image of a person we provide for real-time lip-syncing.

Could you please let me know if this requirement can be fulfilled? If so, what setup is needed, and what are the technical requirements or limitations?