App development

[login to view URL] is an example of a avatar that is speaking, but it is limited to the consumer answering not with voice but by typing on video. We want to allow the consumer to speak in to video, we know we can do the speech to text, as we already do that on Telephone so we can hook in via sip to listen and then make the bot answer, but we don't know how to get this avatar character in as a video participant, allowing us to control the lip movement we do via api. So we thought maybe a screen share, where it is a website which auto plays the avatar or if not, we would have to send over mp4 video files to the solution for each prompt the avatar says. [login to view URL] see this type in a question and see how it responds, we want this same thing, but not have consumer type a question, we will listen via the video call and transcribe what they say via api (we have AI already set up to do this) and be able ot make that avatar respond, just as it does when you type in a msg. So if you envision this avatar maybe being on our server with all white background and on a domain, where it is just sitting there, no text field or anything, we can make that appear as a participant in the video call? if so, we can make the avatar speak from our side, but audio from the webpage would need to piped in to video call. This is one way maybe to accomplish what we seek, or another idea was if that is difficult, we can create a small mp4 video file to stream for each response we need avatar to say. So one video the avatar says the first prompt you see on this video