Voice-driven, talking, animated 3D avatar with Rasa

Hello all,

I have an Android app (called AI Expert) that turns a conversational AI (like DialogFlow) into a talking, 3D avatar automatically; you connect the output from your conversational AI engine to the platform which automatically animates, speaks and gestures through a custom 3D avatar, which you can also put into A/R, or for use as a talking kiosk, etc.

Is there enough interest in this community for me to add Rasa connectivity? The idea would be to specify your Rasa credentials, then the avatar would connect to the Rasa backend, and be powered by that agent automatically.

You can see some examples of the app from this playlist here: https://www.youtube.com/playlist?list=PLH9960zLq-ljbgQzEjK7dsgMMfh22NM1K

and here is the current documentation: https://embodydigital.com/expertaidocumentation/


Ari Shapiro


Hey Ari, This would be pretty awesome if you added it to Rasa!! I am super interested!!!

Ok, great! Is there a standard way to access a Rasa server? Or is it better to have the Rasa agent running on the mobile device?

Awesome - would add a new level of interactivity


Hi @arishapiro!

This sounds awesome, is this project opensource? If so could you please point me to the repo? I would love to contribute.

Thanks for offering.

No, the project isn’t open source, it’s from my company, embodydigital.com . However, the app is free to download and use and let’s you quickly answer the questions: what would my chatbot look like if it was embodied? What would it look like in A/R?

However, if there were some open source Java code that I could integrate that handled (1) authenticating to a RASA server, and (2) making a query and receiving a response from a RASA server, that would certainly speed up the process of getting RASA connected in the app.


Ok, the latest version (2.23) of the AI Expert app now has RASA support.

Feedback from the RASA community would be great.

Here’s a link to the app again:


Does it only work on Android? I am looking for something that will output on a frame buffer on a server so I can support voice and video over WebRTC.

This would be awesome especially if it could be used offline

Hello, is this AI Expert app still works?