Forum SettingsEpisode Information
Forums
Sword Art Online (light novel)
Available on Manga Store
New
Nov 7, 2022 12:01 PM
#1
Offline
Nov 2022
10
So I've been thinking about how the Full Dive system could be made in real life.

So, if I remember correctly, the user's muscles don't function in real life during a Dive, as the signal from the brain goes directly to the game and vice versa (which is why for example blind people would be able to see in VR, if only their eyes are damaged). I can't seem to understand how the real life voice of the user gets transferred to the game, as using your voice would require the usage of muscles. Your brain doesn't determine the basic pitch and sound of your voice (though you can change the pitch and sound of your voice, which would still make use of muscles). I thought maybe you'd need to give a speech sample when you first activate your account, but this way the voice would either be monotone, or you would need to give a lot of samples with different pitches and emotions.

I wonder if anyone knows an answer to how the game copies real life voice to the game. Thanks in advance for a response.
Nov 7, 2022 12:20 PM
#2
Offline
Dec 2020
3
Maybe when you set up the VR you need to add your voice tone
Nov 7, 2022 12:25 PM
#3
Offline
Feb 2021
224
the player uses voice to dive in (link start command). it might just use that to determine the voice.
Nov 7, 2022 1:21 PM
#4
Offline
Sep 2020
632
PoleLord said:
So I've been thinking about how the Full Dive system could be made in real life.

So, if I remember correctly, the user's muscles don't function in real life during a Dive, as the signal from the brain goes directly to the game and vice versa (which is why for example blind people would be able to see in VR, if only their eyes are damaged). I can't seem to understand how the real life voice of the user gets transferred to the game, as using your voice would require the usage of muscles. Your brain doesn't determine the basic pitch and sound of your voice (though you can change the pitch and sound of your voice, which would still make use of muscles). I thought maybe you'd need to give a speech sample when you first activate your account, but this way the voice would either be monotone, or you would need to give a lot of samples with different pitches and emotions.

I wonder if anyone knows an answer to how the game copies real life voice to the game. Thanks in advance for a response.

ai is already pretty good at copying someone's voice just from a short sample. like with deepfakes. though deepfakes use a couple more samples you can already create a deepfake with a really good voice from just a 15 second clip. this is still future tech so it will develop even more and so, you get the nervegear with realistic viice
Nov 7, 2022 1:25 PM
#5
Offline
Nov 2022
10
Spoopy69 said:
PoleLord said:
So I've been thinking about how the Full Dive system could be made in real life.

So, if I remember correctly, the user's muscles don't function in real life during a Dive, as the signal from the brain goes directly to the game and vice versa (which is why for example blind people would be able to see in VR, if only their eyes are damaged). I can't seem to understand how the real life voice of the user gets transferred to the game, as using your voice would require the usage of muscles. Your brain doesn't determine the basic pitch and sound of your voice (though you can change the pitch and sound of your voice, which would still make use of muscles). I thought maybe you'd need to give a speech sample when you first activate your account, but this way the voice would either be monotone, or you would need to give a lot of samples with different pitches and emotions.

I wonder if anyone knows an answer to how the game copies real life voice to the game. Thanks in advance for a response.

ai is already pretty good at copying someone's voice just from a short sample. like with deepfakes. though deepfakes use a couple more samples you can already create a deepfake with a really good voice from just a 15 second clip. this is still future tech so it will develop even more and so, you get the nervegear with realistic viice

Oh right, I forgot about how much it has improved. So do you think that Full Dives could be used for games in the future? And why don't we have it yet? I've seen multiple people use brain activity detection to play games. Can't we just link our brain to a digital brain that copies brain activity and link that digital brain to a digital body?
Nov 7, 2022 1:27 PM
#6
Offline
Sep 2020
632
PoleLord said:
Spoopy69 said:

ai is already pretty good at copying someone's voice just from a short sample. like with deepfakes. though deepfakes use a couple more samples you can already create a deepfake with a really good voice from just a 15 second clip. this is still future tech so it will develop even more and so, you get the nervegear with realistic viice

Oh right, I forgot about how much it has improved. So do you think that Full Dives could be used for games in the future? And why don't we have it yet? I've seen multiple people use brain activity detection to play games. Can't we just link our brain to a digital brain that copies brain activity and link that digital brain to a digital body?

elon musk and another company is working on it. big concern is safety. elon musk has managed to get it working but with inserting somthing into the brain our nervous system(don't remember exactly) and the other company is trying to do it completely external but has not completely succeeded yet but their progress seems promising. i think we can expect some first prototypes or first previews to come out in a year or 3


take this with a grain of salt this is what i remember from some research i did a year ago
Nov 7, 2022 2:18 PM
#7
Offline
May 2021
2
Sorry for my bad english.

Here's a semi game developer's opinion.

I guess one of the possible answers could be that the game has some tweaking parameters that are adjusted by the voice's muscles signals. It could use the audio sample recorded during the setup only as the audio to tweak before the output.

Also about the whispered voice there could be installed a white sound and that could be equalised and pithed based on an analysis of the voice sample, making a personalized whispered voice. This could be interpolated with the main voice based on the signal of the muscle that makes us whisper.

About the shouted voice, I can only imagine it could be done by making a shouting parameter that makes the output more distorted and/or rises the output volume based on some other muscles signals.

What makes it so hard to be done is that all of these information, adding the graphics informations and a lot of other stuff, cannot be done in real time without a big performance loss.

More topics from this board

Poll: » Sword Art Online Episode 21 Discussion ( 1 2 3 4 5 ... Last Page )

Stark700 - Nov 24, 2012

544 by Kraum »»
May 26, 11:59 PM

Poll: » Sword Art Online Episode 6 Discussion ( 1 2 3 4 5 ... Last Page )

Stark700 - Aug 11, 2012

491 by spillkongen »»
May 26, 11:48 AM

» Overall do you enjoy watching SAO? What do you like about it? :) ( 1 2 )

Rinrinka - Apr 15

72 by katnissmaezie »»
May 16, 1:05 AM

» What even is SAO anymore? ( 1 2 )

Jason-Skyrunner - Apr 27

53 by soli1239 »»
May 6, 8:27 PM

» We need a reboot

Superninjaboy2 - Mar 15

32 by Jason-Skyrunner »»
May 1, 5:20 PM

Preview MangaManga Store

It’s time to ditch the text file.
Keep track of your anime easily by creating your own list.
Sign Up Login