Published on 2020-11-11 00:00:00 +0000
I tried to more thouroghly test to see why the problems from last week were arrising. First I turned the visibility of my app to private, in case that affects the voice-to-text stage.
Since all my previous attack commands had been of the form “
I also spent some time looking into what Web Captioner uses as suggested by Prof. Austin Yarger. They use Web Speech API for their voice-to-text. However, it isn’t like a normal API. It is the type of API that is ONLY usable by browsers (and only certain ones at that), much like setTimeout, setInterval, etc. Therefore I don’t believe it is of any use to me, though in theory it could be used in a webgl build (not exactly sure how though).
Attack commands of the form “
At this point I’m not sure what to do regarding voice commands. Should I stick with wit.ai despite the failings as there are no other free crossplatform voice-to-text systems? Should I switch it to use Windows speech which supposedly works better, but would limit the game to Windows only? As I struggled with this question, I continued to work on attack scripts which are independant of the voice system, as well as adjusting the forward of the name bubbles to be the camera position.