The personal blog of game designer, programmer and Cand. Scient. Informatics, Jimmy Marcus Larsen. Currently working as a game designer at Cego ApS.
October 28, 2005 at 21:10
Guess I’m not very good at remembering to write here.
Anyway, I’m here now to deliver a small update to the status of the project. I have not been coding much, but the main game framework is done including different graphics and AI routines. Just need to put together the game logic and that minor speech recognition detail… Most of my time have been used on courses and writing theory for the project. I Developed a clever linguistic theory about computer games design, and is planning to put that to good use soon. The obvious result is that the early design concept might change very much - hopefully for the better.
I also realized that speech recognition is an immature technique, so incorporating its weaknesses (low precision recognition) in the game might be a good idea. But how do I do that? Soldiers who can’t hear my commands because of radio static? Soccer players who can’t hear everything their coach is screaming at them? I have a few ideas, but none of them really are great. Feel free to share your ideas!
October 8, 2005 at 20:00
I have been playing around with the HTK Toolkit, and even though it’s really nice, and well documented as well, I don’t think it will be portable to the Nintendo DS. It will be too slow because of its HMM approach to recognition, so any recognizer of that kind have been dropped. The goal is now to write my own speaker dependant recognizer. Keywords are: speech period detection, dynamic time warp (DTW) and a lot of hard work. The last weeks have also been busy with reading theory on speech recognition, languages and everything else relating to the project.
At the same time, with Christmas getting near, game releases are accelerating so my job as a game reviewer is taking some of my time. I make sure giving the project highest priority as I don’t wan’t to get behind with it, but reviewing games is actually helping it too. Analysing other peoples games can be very rewarding in terms of knowledge of the market, knowing what works and doesn’t and so on. My own design will likely benefit from this.
You are browsing the Game Design Chronicles weblog archives for October, 2005.