Hi Johann ,
Yes that is a starting point, but it can be so much more than that. Our implementation involves some deep learning algorithm modifications coupled with an adaptive and predictive GUI. Speech recognition in Unreal Engine 4 is already available. UE4 is C++, so we have had to recruit a Java developer. Such features are ideal for what Epic Games has currently put out there, so take a look at their new VR Editor
It needs some work, and we would like to add some needed functionality. It was through CBC Spark that we heard of 'Lucida'.
We will be modifying the following section of code:
To gain access to UE4's source code, check out:
We have also been experimenting with IBM's Watson, but we are somewhat drawn to open source!