Real Time AI Demo

Here we show what the reality is really like when using generative AI for smart avatars during a time when OpenAI servers are busy. No edits or cuts here just raw unfiltered video capture.

Generating real voices adds a 4 to 6 second delay on avatar response times when servers are busy however I do believe it’s worth it as you feel as though you are talking to a real person. Normally the delay is approximately 2 to 4 seconds when servers are less active. For comparison, at the end of the video when I speak to our own Athena AI, you will see we don’t generate a real voice for her and the response time is much improved.

Also, for a change I have decided to film this in my Meta Quest 3 using ENGAGE XR’s Mixed Reality / Spatial Computing passthrough. This is ideal for when educators want to bring an historic character into the classroom.

Get in touch if you need some help with AI, Spatial Computing, #Metaverse or if you are in need of immersive technologies for collaboration, training and development.

Find out more here