Discussions

We are developing an interactive digital kiosk where a virtual assistant (a human avatar) will respond to user queries in real-time. We are considering D-ID’s real-time streaming API for lip-synced avatar responses and would like to confirm if your solution meets our specific requirements.
Project Overview

A pre-recorded human avatar (with multiple angles/images) will be used to generate real-time lip-synced responses.

Users will interact via voice, and the avatar should respond with synchronized facial movements and speech.

The responses will be from a predefined FAQ database 
Later, we may integrate LLM-based dynamic responses 

The avatar must support multiple languages.