By using this site, you agree to our Privacy Policy and our Terms of Use. Close
timmah on 07 January 2013
Edit Thread
Mazty said:
Mnementh said:
Mazty said:
 


If we run off the voice-to-text software, that's using a dual core 2.2ghz to just write it down. To analyse the sentence as well would be too taxing on the Wii U. It could be done, but I'm doubtful that it could be done in a reasonable amount of time on the wii u as in you don't want a 10 second pause after every command.

And why do you think a game would implement a full voice-recognition needed for dictating texts and an analyzer for sentence-structure? Most voice-recognition only recognizes some words and that would be mostly enough for a game if used clever. nobody want's to port Dragon Naturally Speaking to console.

But to even cherry pick words whilst doing anything else would still be very demanding. It'd have to be done in a "safe" (read no AI) area e.g. hitman safehouse or the likes. It couldn't be done if anything else was going on. 

If you honestly think that requires a lot of processing, I really don't know what else to tell you. I used to voice control an old (slow) P3 computer running Windows 2000, also, one of the Rainbow Six games on the original Xbox allowed in-game voice control of the squads and that processor was certainly very weak by today's standards. The first voice control system for a GUI came out in the late 80's, so obviously it can be done with VERY limited processing resources. I don't know where you get these ideas, simple voice recognition and control has been around a long time, worked on much, much weaker hardware than what we have today, and does NOT require much processing power. Cherry picking the keywords out of converted text is also not processor intensive at all, neither is a simple text based query to find the propor response or action. We're not talking about something like Siri here.