Advertisement

Inquiry on Voice-Over MMORPG Feature

Started by May 02, 2005 03:14 PM
56 comments, last by hplus0603 19 years, 9 months ago
Quote:
Original post by wackatronic
Everytime a caster learned a new spell, you could have the voice recognition be trained. this way you won't have to worry about pronuciation becuase what you get when trained is how the player pronounciates the spell words.


I like this idea as well. With the voice recognition as part of the client, the trainer or teacher could have the player repeat again and again (the role-playing mask would be that the caster needs to get it right). Once the pattern is recognized the spell will be casted anytime that pattern or incantation is used. We should monitor the length of the spell however, so that lazy players do not simply speak "spell one" where they should be reciting a longer incantation.

Keep the feed-back coming. I appreciate and consider everything that is said here.
Quote:
Original post by JJacobo
I like this guy. :D
Thank you for the input, I understand what you're saying about filtering the speech on the speakers client so that it IS an entirely new and encoded piece of information when received by listeners of another race.


Not really but when you have people playing MMORPGs, you usually have a fairly large group of purist that like to play in character, chat in character, and now even speak in character (not by language).

In the case of spoken incantations/spells, the spell that is spoken/uttered is validated through the voice recognition prior to sending any type of command for action (spell effect) to the game server. This will eliminate unecessary bandwidth traffic between the clients and server.

My systems are going to have voice chat and I approached it very similar to the effects you mentioned. As far as echoes and such, I know for sure you can apply this type of filtering through DirectX, but I'm not to sure about OpenAL, but it will more than likely be the same.

Selective filtering is will have to be a main point. By allowing the icon or character to be clicked and then voice chat initiated would be nice. Private voice chats would have to be approved by both parties to ensure griefers do not abuse other players.

I think game speech is a very important emerging technology for MMOs and hopefully people like yourself will help in setting the standards. Good luck.

-- Eric 'Wackatronic' Tomlinson

Fear is the mind killer. But if you have no mind, do you have no fear?
Advertisement
I appreciate it very much. I'm be sure to keep the Gamedev.net community updated on our progress, though you'll more likely be hearing from me looking for talent or asking for help!
=P
Quote:
Original post by wackatronic
Quote:
Original post by Raduprv
The problem is, if you want to apply filters and cast spells by speaking the words, the server will have to receive the sound, process it (see if it's a spell), add the appropriate filter for EVERY player in the range, then send the modified sound (each player hears it differently, depending on race, language skill, etc.) to the players in the range.
Now, unless you require broad band as a requirement for the players (ie. no dialup and ISDN) and have a LOT of bandwidth on your server, this will be impossible.<!–QUOTE–></td></tr></table></blockquote><!–/QUOTE–><!–ENDQUOTE–><br><br><br>Have the voice recognition &#111;n the client, it processes and validates prior to sending the command to the server for action. This way it never leaves the client until it has been recognized as a valid incantation.<br><br>– Eric 'Wackatronic' Tomlinson<!–QUOTE–></td></tr></table></blockquote><!–/QUOTE–><!–ENDQUOTE–><br><br>The problem with this is that people would automatically look to hack the client so that they can bypass the voice recognition and go straight to the command. That way the skill of speaking would be a complete waste because the player instantly attains perfection. Generally anything you do &#111;n the client can and depending &#111;n the popularity will be hacked. Thats why common practice is to make MMORPG clients more like dumb terminals instead of decision makers, because it is incredibly difficult/impossible to make those client processes secure.<br>
Turring Machines are better than C++ any day ^_~
Well perhaps a series of checks could be put into play. As I mentioned players will have no sort of command line control, but as for them hacking the client itself we may have one method of determining whether a player is engaging in the behavior.

Since all spoken words, be it a spell or a "I would like to BUY" command to an NPC, will not only be processed and broken into commands by the client but heard by all surrounding players.

Therefore, it might be easier to detect these cheaters by making sure that the stream duration or tone or such for a spell or command is within the appopriate range allotted to it. If this was enforced, it would be silly for players to speak something phoney for approx. the correct duration only to use a client tweak/hack and risk being banned for it.

Whether we could actually be checking this from the server or perhaps set up a player watch dog, type of system remains to be discussed in detail. =P

Any thoughts?
You wouldn't have to ever speak if you didn't wish too, but taking on skills in casting would be impossible for you. Surprisingly, from the focus groups we have done with current MMORPG users, a good majority of them that have used some sort of voice program wouldn't play without it and most of those who haven't have seemed very open to the idea, so long as the system isn't too complicated. If you read the entire thread though, you'll see that typing to communicate is an option. Moreso, each of the other features of the game have a typing equivilant. It is only spellcasting that requires speech.

I do see your criticism though, but encourage you to read all of my posts. I think that if it WERE required, your thought would apply, since not everyone has the self-esteem to speak out loud and into a game world.
Advertisement
Quote:
Original post by Anonymous Poster
Well I assumed since you were talking about preventing the player from bypassing the voice casting and such that it would be required, otherwise there is no point in any security measures that you have been talking about for the last part of the post.

For casting, the voice system IS required. But for communicating with others, or interacting with NPCS it is optional, since we understand that some players may feel uncomfortable or simply have to be quiet temporarily. The speech is still heard audibly though, in the pre-set speech voice you select at character creation. We do 100% force casters to have to speak, if that is what you are still confused about. We think that casting should require a unique type of challenge with it, and the incantations and dynamic aiming systems for casters fulfill that goal. If you think that it should be otherwise we're more than open for discussion, but I wanted to clear that up. =P
Thanks for feed-back btw.
There's no need to get testy, I am listening to your opinion and considering everything you're saying. However, I think it is worth mentioning that unlike most MMORPG's casters will not be constantly using their spells unless they are indeed in combat and even then the amount of incantations is limited like this:

Lets say the spell for Flamed Missles (not real names here), was "Incantum Desunis Rinaldes" (again Gibberish.) As a caster you might line up, speak the incantation and from there enter 1st person mode where you would manually laugh the fireballs which would continually spawn in your hand unless you cancel the spell or change to a new one. Thus spells function is a sort of toggling mode and eat up concentration (a hidden pt. system). So If I wanted to heal players I might enter a healing mode by using the enchant (though only one spell would be available). Thus you won't be "constantly talking" or anything of that nature, but memorizing your spells and knowing when to use them will be paramount. Does that make more sense?
Quote:
Original post by intrest86
The problem with this is that people would automatically look to hack the client so that they can bypass the voice recognition and go straight to the command. That way the skill of speaking would be a complete waste because the player instantly attains perfection. Generally anything you do on the client can and depending on the popularity will be hacked. Thats why common practice is to make MMORPG clients more like dumb terminals instead of decision makers, because it is incredibly difficult/impossible to make those client processes secure.


This is why you design in proper packet encryption and validation. In order to hack the client you need the source. If you try to spoof packets, a good token system would reject the tampered packet.

-- Eric 'Wackatronic' Tomlinson
Fear is the mind killer. But if you have no mind, do you have no fear?
I believe that something like this would have to be put into play, since our project requires a lot of crunching, and we will simply need the clients to share the burden.

This topic is closed to new replies.

Advertisement