Rethinking my AI server structure.
At the moment if I continue developing it, the traffic flow would be
-Server detects a speech keywords and sends it to UNITY
-UNITY receives the keyword and parses it for logic to manipulate itself (model animations) and create the AI
-UNITY sends it's phrase it wants to synthesize back to the server to be spoken out loud
Now this is fine and all, but I think I am going to change the way to two interact further abstracting which application does what. Right now the server would handle speech recognition and speech synthesis, and UNITY would handle AI logic and model animations.
What I am going to change it to is the server handles all AI logic, speech, and everything else, and it will simply send UNITY data required for the animations. This will allow UNITY to do what it's good at while having the full power of C# to handle AI logic and various other acts.