Bob Rissolo · Dialogue Lead

Oh, yes, yes, yes. So the last couple of locations, I think we've done a much better job. And the first test cases of really improving this were the cargo and refinery decks, which were location-generic, and they really had a great positive reception. The dialogue landscape has a foreground: actual AI interacting, words coming out of the AI's mouth. But then you have, sight unseen, this middle ground, the chatter that they're referring to, that we've really been focusing on. It doesn't have AI, it doesn't have any other stimuli other than that you're in a populated space and you want it to have that immersion and that feeling that there's life there when maybe there's not really anything actually triggering it. And that was a big thing that we did with these spaces. The one last thing is the base, the very background walla, which is that indiscernible, almost sound effect of dialogue, just filling the space. And we've added the layers of that, but it's still this static type thing that's just for these spaces. And we did this, I think, really well with Orison, where we went and picked out these densely populated areas that we knew would usually have that type of population, and we were able to add those chatters so that you had the dialogue even if the AI wasn't there really acting and vocalizing. And we had the walla in the background. So it's these layers compounding together that gives you a dialogue landscape and a feeling of life. But it's all still very much us attacking these spaces and whatnot. And we've taken this, Orison being a very good full prototype of doing that for a whole location, and we're now going back to Lorville and some of these other locations and really trying to fill them out as well. So we've taken this and we're generating more and more, and it's really helping. We've hit the hospitals with some vignette-type chatter. We want to do the habs as well. You're going to find some cool stuff: walking through the halls and hearing some weird stuff, especially in the more fun locations like Grim HEX. Get some cool stuff in there. And Pyro is going to be fun too, coming up with a lot of that outlaw stuff. That's actually where we're hoping to get a lot of the dialogue life, because we won't have as much character content for such a new location, right? So it's going to be really that middle ground and that background that will really help with those spaces. In our future, we would really like to have more of a dynamic walla system that can not just say that this is a space that has population and so we should treat it as such, but actually look at the populations dynamically, because it's not just NPCs roaming around, it's players. And we want to make sure that if you stand in a room with a bunch of players, that actually also feels like a densely populated space. You don't know when that's going to happen. It's going to be very dynamic. It's going to have a mood. It's going to have a population and all those things. We want to have a dynamic of the types of populations in there, and we'll take stems and a lot of sources of walla and mix appropriately to have you feel like that's the space that you're in. And that's going to take a lot of content. It's going to take a lot of tech and a lot of prototyping, a lot of work on it. But we have almost started the process. We're getting there. To be honest, we have some big choices to make with the type of software that we want to use, because part of the big thing that we want to do with creating so many assets is using text-to-speech and speech-to-speech, and using these kinds of tools to help bolster what recordings we already have, and really to help us generate the right amount of content without having to go and record and record a bunch of walla sessions. So we're looking at that tech right now, and it's quite a substantial monetary decision to make. So we need to do our homework. We need to do our research. We need to compare and weigh against and all those things. So we're at that level at the moment where we're about to engage with some good research. And then once we pick that, then we can really start to move on with getting a proper walla system in there. So that's, I'm sorry, I kind of blabbered on there.