Would an open world game benefit from voice recognition?

Sorry, but that question is far too vague to be answered in a meaningful way.

Sorry, but that question is far too vague to be answered in a meaningful way.

Ok.
For instance, the cutscenes in games. Would it be good if you had to actually speak to whoever was in the cutscene. What happens after that depends on what you said during the cutscence.
Would you like to negotiate payment for missions by speaking to whoever was giving the mission and depending on your tone, the payment could be reduced or increased.
On a mission say a bank robbery, you tone determines the outcome. If you yell at the hostages, they stay in line but if you sound jittery and scared, they could fight back.

It would obviously be cool to interact on that level with game NPCs, but I am not sure the work would equate to better fun factor. The issue I see is that if players can talk to NPCs, they will want meaningful conversation with them, real choices, expressive NPC AI, etc. And that is an entire gigantic project on its own, without any real game at all. Taken together with a large, open world game, we are probably talking about a giant project for a 100+ man team.

And anytime you bring in a new input channel to games, there are always problems. Look at many of the games on Xbox 360 that relied heavily on Kinect for input. Many of them got terrible ratings amid strings of complains of recognition issues. Players want razor-sharp controls (that includes all forms of input) all of the time, and sadly, this has dampened the reception to a lot of innovative input devices.

It would obviously be cool to interact on that level with game NPCs, but I am not sure the work would equate to better fun factor. The issue I see is that if players can talk to NPCs, they will want meaningful conversation with them, real choices, expressive NPC AI, etc. And that is an entire gigantic project on its own, without any real game at all. Taken together with a large, open world game, we are probably talking about a giant project for a 100+ man team.

And anytime you bring in a new input channel to games, there are always problems. Look at many of the games on Xbox 360 that relied heavily on Kinect for input. Many of them got terrible ratings amid strings of complains of recognition issues. Players want razor-sharp controls (that includes all forms of input) all of the time, and sadly, this has dampened the reception to a lot of innovative input devices.

It will be cool, take a good amount of effort and if it isn't almost perfect, the consumers don't care about your effort.

I think there's practical reasons why it would put some players off, no matter how cool an effect it is. I have a six month old son asleep upstairs and my wife's trying to do her college assignment on the sofa next to me: the last thing I want to do right now is scream at a hostage! I'd require some alternative input mechanism to carry out equivalent functionality, otherwise I simply wouldn't be able to play your game.

I think there's practical reasons why it would put some players off, no matter how cool an effect it is. I have a six month old son asleep upstairs and my wife's trying to do her college assignment on the sofa next to me: the last thing I want to do right now is scream at a hostage! I'd require some alternative input mechanism to carry out equivalent functionality, otherwise I simply wouldn't be able to play your game.

lol. That's one of the weird things of voice recognition.
Imagine shouting get down on the floor! to a computer.
It is still a project for me, thus the signa . . .

It's obvious an innovatibe and cool idea. The gameplay could benefit from the voice recognition, but not too much. You will probably not get it working too great, especially if you see large companies like Google and Apple still have a lot of mistakes... Let alone multiple languages, people who don't want to shout to their PC or who can't because of e.g. one of the reasons mentioned above.

Using an existing tool would help, but there is no aceptable one yet, in my opinion then. I don't want the PC to understand they have to kill the hostages while I wanted to said they should protect them... But I can't really judge because my native language isn't English, like many other people...

Even though it may not be practical, it could be cool for a tech demo! ;)

Localization would be a nightmare.

It's obvious an innovatibe and cool idea. The gameplay could benefit from the voice recognition, but not too much. You will probably not get it working too great, especially if you see large companies like Google and Apple still have a lot of mistakes... Let alone multiple languages, people who don't want to shout to their PC or who can't because of e.g. one of the reasons mentioned above.
Using an existing tool would help, but there is no aceptable one yet, in my opinion then. I don't want the PC to understand they have to kill the hostages while I wanted to said they should protect them... But I can't really judge because my native language isn't English, like many other people...
Even though it may not be practical, it could be cool for a tech demo! ;)

I was just wondering how to make games more realistic.

I would make the tool rather than using a tool and include it in my engine