What I'm also looking forward to is a better and more real sounding speech synthesizer. We don't use the voice actors' faces for the games a lot and using these synthesizers will eliminate them from the equation completely.
This is a great point. Speech Synthesis is something that's kinda like the Flying Car to me. When I was younger, it seemed like a no-brainer that we'll have it in the near future. You could already see crude examples. So making that leap made sense.
How come we don't have it yet? Kinda boggles my mind.
In fact its not mainly fps we detect, its the sudden changes in fps that we notice. You can get used to 30fps, and not notice any problem with it. Of course higher is always better, but 30 is adequate in most cases as long as it doesnt fluctuate too much.
This is a great point. Speech Synthesis is something that's kinda like the Flying Car to me. When I was younger, it seemed like a no-brainer that we'll have it in the near future. You could already see crude examples. So making that leap made sense.
How come we don't have it yet? Kinda boggles my mind.
Actually, it has gotten better over the years(http://www.acapela-group.com/text-to-speech-interactive-demo.html). It's just that it still sounds synthesized and there is no need for people to use it if they can use actors.
Since there isn't enough focus on it, like computer graphics, it's not improving as quickly. I guess it was the same thing with computer graphics during the 80s but as soon as they saw more economical benefits(Jurassic park) they started using it more and improving it quicker.
My thoughts are that a speech synthesizer would work great for an AI with good speech abilities which's dialogue would depend on the player's reactions and so on. It would actually come to life!
In fact its not mainly fps we detect, its the sudden changes in fps that we notice. You can get used to 30fps, and not notice any problem with it. Of course higher is always better, but 30 is adequate in most cases as long as it doesnt fluctuate too much.
That might be the case for platformers or 3rd person shooters, but for an fps game 60fps makes a huge amount of difference. Playing Quake 3 at 30fps is not a fun experience.
Actually, it has gotten better over the years(http://www.acapela-group.com/text-to-speech-interactive-demo.html). It's just that it still sounds synthesized and there is no need for people to use it if they can use actors.
Since there isn't enough focus on it, like computer graphics, it's not improving as quickly. I guess it was the same thing with computer graphics during the 80s but as soon as they saw more economical benefits(Jurassic park) they started using it more and improving it quicker.
My thoughts are that a speech synthesizer would work great for an AI with good speech abilities which's dialogue would depend on the player's reactions and so on. It would actually come to life!
Also it would be easier to make changes and additions to the scripts, like the old days where you didn't have voiceover for all conversations. Its one of the reasons that you got limited conversation topics in Oblivion, compared to Morrowind, because they wanted it all spoken out loud.
And I know from personal experience, that when there is an error in audio or something isn't quite right, its very likely its never going to get fixed, because they have already recorded audio and wont be doing it again. Would be much better if you could just add some now text lines and its all fixed automatically. (yes I know its more tricky then that).
kind of serious...
Wouldn't it be amazing if the realtime cinematic software 2020 roadmap was ushered in with a phone that output a wireless 1600p res. ( even better would be a brite projector phone with a sharp contrast ratio and brightness, but I doubt either that or a true broadband revolution will happen in my lifetime as in 8 days I will be closer to 50 than 40 and my family is cruely taking delight in counting down :poly118: )
That Doom 3 does have a nice psycho shadow casting. without any soft edge solution they can cast some batman angles that would explode into acne/biasing hell for me. I use to play Doom 3 by myself late at night in my warehouse and I would be absolutely terrified.
The cost benefit to synthesized speech seems obvious to me. If you have a game like that new Star Wars MMO where everything is voiced, I'm sure it would be a hell of a lot cheaper to use a synthesized version. Not to mention that you'd be able to do a lot more with it.
But it's purely the fact that the quality is so poor. Though I agree, it does sound a lot better, I'm yet to hear one that could actually pass for a person.
Yeah, it sounds so cheap that they won't even use it for computers in games(portal).
Imagine an mmo with integrated text-to-speech for all the players. Imagine cities filled with players talking to each other, it would add so much to the atmosphere! @Mcgreed: Exactly, and the cost for using it would be so much less than hiring some overpaid hollywood actor/actress who probably doesn't even have any experience in working with voice acting.
What I'd like to see with that actually, now that I think about it, is some fantasy game where the characters will use it to speak an in-game language. Sure, you could tell it's not a person speaking English with it. But if they spoke some weird fantasy language, then who would be the wiser?
That'll be a good way to hone it until it's good enough to move into human-intelligible language. Hell, as is, if they made it speak Chinese or something, I doubt I'd notice it all that much.
(BTW, anyone remember the old Goblins games? They solved that problem really elegantly there I thought)
I think/would hope the Smaritan Demo will be possible on the next consoles. By then it will definitely be possible on PC probably with a single GPU instead of 3.
@Bigjohn: I think the problem is when people can tell that it's synthesized speech. It's really well done but you can tell it's not real. I've heard people with no knowledge whatsoever of vfx say "a visual effect which you can tell is a visual effect is a bad visual effect". It's not true per se, since the problem usually lies in the effect not fitting into the recorded film and not the effect itself, but it just ruins the atmosphere. Don't get me wrong, there are a lot of things that can currently ruin the atmosphere such as repetitive sound clips, but I guess a few sound clips with jiberish comes off cheaper than an entire system for the speech.
Right, which is why I wouldn't want it (currently) used instead of English. But like in that Goblins example, it's still within your suspension of disbelief that you're in a Goblin world. So you just assume they speak Goblin. Worked for me when I was a kid even with that very limited vocabulary.
But then looking at modern fantasy worlds, where they actually come up with some pseudo-language for those creatures, I think it could work.
But I see your point. If it comes off as being "fake", it'll break the immersion. Who knows? I just really want to see that field progress in the near future.
Replies
This is a great point. Speech Synthesis is something that's kinda like the Flying Car to me. When I was younger, it seemed like a no-brainer that we'll have it in the near future. You could already see crude examples. So making that leap made sense.
How come we don't have it yet? Kinda boggles my mind.
Actually, it has gotten better over the years(http://www.acapela-group.com/text-to-speech-interactive-demo.html). It's just that it still sounds synthesized and there is no need for people to use it if they can use actors.
Since there isn't enough focus on it, like computer graphics, it's not improving as quickly. I guess it was the same thing with computer graphics during the 80s but as soon as they saw more economical benefits(Jurassic park) they started using it more and improving it quicker.
My thoughts are that a speech synthesizer would work great for an AI with good speech abilities which's dialogue would depend on the player's reactions and so on. It would actually come to life!
That might be the case for platformers or 3rd person shooters, but for an fps game 60fps makes a huge amount of difference. Playing Quake 3 at 30fps is not a fun experience.
Also it would be easier to make changes and additions to the scripts, like the old days where you didn't have voiceover for all conversations. Its one of the reasons that you got limited conversation topics in Oblivion, compared to Morrowind, because they wanted it all spoken out loud.
And I know from personal experience, that when there is an error in audio or something isn't quite right, its very likely its never going to get fixed, because they have already recorded audio and wont be doing it again. Would be much better if you could just add some now text lines and its all fixed automatically. (yes I know its more tricky then that).
kind of serious...
Wouldn't it be amazing if the realtime cinematic software 2020 roadmap was ushered in with a phone that output a wireless 1600p res. ( even better would be a brite projector phone with a sharp contrast ratio and brightness, but I doubt either that or a true broadband revolution will happen in my lifetime as in 8 days I will be closer to 50 than 40 and my family is cruely taking delight in counting down :poly118: )
That Doom 3 does have a nice psycho shadow casting. without any soft edge solution they can cast some batman angles that would explode into acne/biasing hell for me. I use to play Doom 3 by myself late at night in my warehouse and I would be absolutely terrified.
But it's purely the fact that the quality is so poor. Though I agree, it does sound a lot better, I'm yet to hear one that could actually pass for a person.
Imagine an mmo with integrated text-to-speech for all the players. Imagine cities filled with players talking to each other, it would add so much to the atmosphere!
@Mcgreed: Exactly, and the cost for using it would be so much less than hiring some overpaid hollywood actor/actress who probably doesn't even have any experience in working with voice acting.
That'll be a good way to hone it until it's good enough to move into human-intelligible language. Hell, as is, if they made it speak Chinese or something, I doubt I'd notice it all that much.
(BTW, anyone remember the old Goblins games? They solved that problem really elegantly there I thought)
Edit:
[ame="
Just instead of such a limited "vocabulary", come up with something legit.
What do you guys think about raytracing? coming anytime soon?
But then looking at modern fantasy worlds, where they actually come up with some pseudo-language for those creatures, I think it could work.
But I see your point. If it comes off as being "fake", it'll break the immersion. Who knows? I just really want to see that field progress in the near future.