I was thinking of my next game project(I know I should finish the first before planing the second, but some times you have a good idea and if you don't add it to the plan you might forget it.) as well as anime, fighting games and voice communication. And an idea popped into my head why not use voice recognization for the fighting as aspect of an anime style game? Now anyone who's seen an anime will have noticed they tend to call out the names of their moves when they perform them so why not incorpate that into a game? Instead of selecting the "ULTRA MEGA INFERNO FIST" move from a skill list you could instead call out "ULTRA MEGA INFERNO FIST" into the mic to perform the move. Would that be fun and amusing and keeping with the tone of the game? You don't need use this feature to play the game but it is more immersive and amusing to use it. So in general how do people feel about game features that are are essentially purely ornimental? Do they add too a games enjoyment or are they waste of development resources? Some possible ornimental features for my current game could be: 1) Able to listen to your favorite mp3s in game instead of the standard music. 2) Able to add your own GUI themes, allowing you to change any of the in game colors and graphics. 3) ...

Hrm... I just wouldn't want to get kicked in the head while I was busy shouting some long senseless name.

In an anime, the action tends to pause while one of the fighters screams something like this. Some fighting games simulate this so that when you perform some special move, the action stops while you float in the air or whatever, and then the attack is performed. Your opponent is held or paused as well. This makes more sense when you issue the command through a sequence on your controller, because it's faster, but I don't know what it'll be like through a voice command.

Take a look at the DragonBallZ fighting games. The most recent one, Budokai 3, is a lot of fun, even though I have no idea what's going on, story-wise. It has dramatic pauses in the action when particularly devastating moves are executed, and some nifty cinematics accompany them. They aren't just fluff, since the defending player has a chance to avert/mitigate their effects, but if you used a system like that, and you could earn a "moment of advantage" at which to execute a voice-activated move, it could work well.

Keep in mind the current unreliability of voice recognition - it's a lot better than it used to be, but still makes plenty of mistakes - I doubt gamers would be very happy if they failed to pull of a move due to faulty voice recognition, or even worse, compromised theselves due to the wrong move being executed.

current voice recognition sucks when you're trying to interpret general commands, but if you were to create a profile and train on the move list (say, 3 times per move), it should be pretty accurate. Look at MS Game commander and other similar software.

This sounds interesting. You could develop a vocal combo system, you pause the action while one players chants his attack, you could try a lot of stuff. Sounds pretty cool to me.
Training is a must, specially for foreign players with an accent =)

And for god's sake, change the topic's name, it says nothing to me =D

No matter whether there is good voice recognition or not, I would not want a voice command because it is tiring, eventhough you just use it once in a while. Voice command is not really that fun, it is like talking to a kid, telling him what to do. Besides, movies and animes uses dramatic presentations. In other words, they cheat. The characters know exactly when they are going to get hit. Suppose you make a very strong move to be voice-commanded, such as 'kamehameha', in an ideal world, the player would shout at the dramatic moment perfectly to activiate the critical attack. However, in reality, the player would have to wait for the right opportunity to use the attack, and say, "kamehame---------------ha!" instead.

That was the subjective thoughts and speculations. Objectively:

When you check the interfaces of real-time games, one prominent feature is macroing. Macros allow complicated actions be done with a single hit of a key. It is a desirable feature because it allows the player to have precise control. Voice commands go against this precise control.

For the ornamental features 1) and 2), if you game is big enough and you use scripts, the player will have ways to achieve those (i.e. they can mod it). Consciously including those features is no more than making an interface. There is not much waste and not much gain for adding them, because you can always add those features afterwards if you want to. Unless it is Dance Dance Revolution, where adding a song means adding a new playable level. But that is not ornamental.

Okay, So I'm seeomg a strong anti-voice control sentiment, and the game idea I was thiinking of using this in isn't a fighting game itss a sim with some combat elements. So bare that in mind and that you wouldn't need to use it but you could if you wanted. Are you telling me that you would never use it? Forget about the mechanics, and timing, and all other issues, if you had the ability invoke moves by calling out their names in an anime gaming would you? And if there was some kind of bonus applied to the move when using voice commands would it make it more worth wild? Or would you feel cheated when you didn't use it?

thoughts:
when you are making a sharp turn in a console racing game, it seems to help if your body also turns along with the controller. Should those who turn physically gain a bonus effect in the race?

your idea of voice command is getting away from being ornamental. if you compare that idea with the thought above, you are creating a reward effect for players who are immersed in the game. If you simply want to have this reward but keep it ornamental, then I would suggest that the voice command only activate a dramatic cutscene.

In other words, when you are using Kamehameha with Goku, there is a time frame in which you can shout 'Kamehameha', which will trigger the dramatic slow motion and extra flashings. There is not bonus damage, but bonus drama, rewarding immersion with immersion.


still, I would rather just hold down an extra button to trigger the dramatic scene, and would rather listen to the voice of the characters than my own voice. (i.e. if I issue the attack while holding down the 'intense' button, instead of the non-interrupted action, the game stalls for the character to shout the move and do the flashes.)

Listening to your own voice might actually break immersion. Maybe instead of using the voice command to command your own character, use it for commanding teammate NPCs. Imagine a game in which you play a evil spirit and the gameplay is about you possessing the NPCs to do your evil deeds:

"sakura, sakura, take the knife... yes, the knife, stab sasuke, stab him, Yes, STAB him!"



If you see voice command as an alternative control, there is not reason to reject it in a perfect world. The issue would be the same as key config, and different people would have different preferences. In terms of the bonus effects of voice commands, you can pretty much do whatever you want if it is a single player game. Some people might see it as a voice-activated cheat code, but a lot of people would find it fun to have cheat on in single player (and more so in multiplayer). It will be fun to have both the mild bonus effects and the outrageous effects.

Another reasonable ornamental feature regarding voice, would be to allow the player customize the voice of the avatar.


[Edited by - Estok on March 7, 2005 8:12:08 PM]

Quote:

Original post by Kazgoroth Keep in mind the current unreliability of voice recognition - it's a lot better than it used to be, but still makes plenty of mistakes - I doubt gamers would be very happy if they failed to pull of a move due to faulty voice recognition, or even worse, compromised theselves due to the wrong move being executed.

I'm trying to figure out how such a system works even reasonably well...All across the world there are different regional dialects, and accents...I'm from Chicago and pronounce certain letter combinations different ways then my girlfriend from North Carolina...then of course there would be problems with people whom have speach disorders, lisps, stuttering, etc...How do voice systems handle such issues? what about homophones (words that sound the same, but have different meanings and spellings)?

Not to mention the fact it would get tireing hearing someone(thinking of children) repeatedly yelling out such nonsensical phrases just to play the game.

For starters they don't have to be nonsensical. If i see another dragonball quote... argh!

anyways. If it were a fighting game you could pause or slow down time, so no cheating on the words timing =D but its a sim game so:
In some sim games, when things get heated up one tends to misplace keys. "Chaff! no thats not the chaff... WHERE'S THE DAMN CHAFF!!" Specially when things are going so fast you can't stop and look at the keyboard. In such cases, just saying "Chaff" and the game launching one, would be a timesaver. I'd consider it just like a keybinding. Like having a good joystick. No shame in that.

About accents and stuff, its really easy (as voice recognition goes). Its just like a key binding. The game goes: say your word for Chaff. Then you say: "kamehameha" It becomes a binding, and now the algorithms compare your voice aganist your own voice, and the matches should be pretty obvious. And when you say it, I shall pop from the darkness and bop you on the head.

Good candidates for voice bindings are weapons, but hey, if you have a key and joystick binding system, why not just toss voice in it too? make anything that can be bound bindable to a custom voice command. Yay.